From f804bbd679557297ef6089d02f5aa9706d0cce04 Mon Sep 17 00:00:00 2001 From: tzhouam Date: Wed, 9 Sep 2026 10:49:37 +0800 Subject: [PATCH] [Edge][WP0] Record Qwen3-TTS streaming baselines on 1x L20X Default deploy config (1 -> 25 frame schedule), 12 prompts x 3 repeats, warm-up excluded, one GPU via `gpu run`: 1.7B CustomVoice: TTFA p50 32.0 ms, playback-start p50 116 ms (stall at TTFA 85 ms), RTF_total 0.086, chunk gap p50 163 ms, init 271 s (cold caches) 0.6B CustomVoice: TTFA p50 27.8 ms, playback-start p50 95 ms (stall 67 ms), RTF_total 0.077, chunk gap p50 146 ms, init 152 s Per-request chunk arrivals, memory and the full summaries are in the JSON (force-added: *.json is gitignored repo-wide). Co-Authored-By: Claude Fable 5.1 --- .../default/20260909-104204_baseline.json | 1592 ++++++++++++++++ .../default/20260909-103504_baseline.json | 1620 +++++++++++++++++ 2 files changed, 3212 insertions(+) create mode 100644 benchmarks/tts/edge_results/stream_latency/Qwen3-TTS-12Hz-0.6B-CustomVoice/default/20260909-104204_baseline.json create mode 100644 benchmarks/tts/edge_results/stream_latency/Qwen3-TTS-12Hz-1.7B-CustomVoice/default/20260909-103504_baseline.json diff --git a/benchmarks/tts/edge_results/stream_latency/Qwen3-TTS-12Hz-0.6B-CustomVoice/default/20260909-104204_baseline.json b/benchmarks/tts/edge_results/stream_latency/Qwen3-TTS-12Hz-0.6B-CustomVoice/default/20260909-104204_baseline.json new file mode 100644 index 000000000..58a675e44 --- /dev/null +++ b/benchmarks/tts/edge_results/stream_latency/Qwen3-TTS-12Hz-0.6B-CustomVoice/default/20260909-104204_baseline.json @@ -0,0 +1,1592 @@ +{ + "schema": 1, + "meta": { + "model": "Qwen/Qwen3-TTS-12Hz-0.6B-CustomVoice", + "deploy_config": null, + "query_type": "CustomVoice", + "txt_prompts": "benchmarks/tts/prompts_12.txt", + "git_sha": "b22dcc59", + "gpu_name": "NVIDIA L20X", + "cuda_visible_devices": "0", + "timestamp": "20260909-104204", + "argv": [ + "--query-type", + "CustomVoice", + "--txt-prompts", + "benchmarks/tts/prompts_12.txt", + "--init-timeout", + "1500", + "--stage-init-timeout", + "900" + ], + "bench_args": { + "model": "Qwen/Qwen3-TTS-12Hz-0.6B-CustomVoice", + "warmup": 1, + "repeat": 3, + "init_only": false, + "tag": "baseline", + "output_dir": "/data/zhoutaichang/embedding_infer/vllm-omni/benchmarks/tts/edge_results/stream_latency", + "save_wav": false, + "gpu_poll_interval_s": 0.2, + "parallel_stage_init": false, + "deploy_profile": null, + "bench_help": false + }, + "warmup": 1, + "repeat": 3, + "init_only": false + }, + "init": { + "init_s": 151.48892775084823, + "init_timeline": null + }, + "resources": { + "peak_rss_bytes": 10019835904, + "peak_gpu_mem_mib": 46846, + "gpu_mem_samples": 204, + "gpu_mem_mode": "per-process" + }, + "requests": [ + { + "request_id": "w0", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 474.6216149069369, + "gap_ms": 474.6216149069369, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 623.8525877706707, + "gap_ms": 149.23097286373377, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 766.3595001213253, + "gap_ms": 142.5069123506546, + "samples": 48000 + } + ], + "t_final_ms": 808.134017046541, + "ttfa_ms": 474.6216149069369, + "playback_start_ms": 543.8525877706707, + "stall_at_ttfa_ms": 69.23097286373377, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.56, + "total_wall_s": 0.808134017046541, + "rtf_streamed": 0.1878332108140503, + "rtf_total": 0.17416681401865106, + "gap_ms": [ + 149.23097286373377, + 142.5069123506546 + ], + "text": "Hello, welcome to the voice synthesis benchmark test.", + "warmup": true + }, + { + "request_id": "r0_0", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 26.67097607627511, + "gap_ms": 26.67097607627511, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 179.53822435811162, + "gap_ms": 152.8672482818365, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 321.8691791407764, + "gap_ms": 142.33095478266478, + "samples": 48000 + } + ], + "t_final_ms": 354.7675833106041, + "ttfa_ms": 26.67097607627511, + "playback_start_ms": 99.53822435811162, + "stall_at_ttfa_ms": 72.86724828183651, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.48, + "total_wall_s": 0.3547675833106041, + "rtf_streamed": 0.07888950469136677, + "rtf_total": 0.0777999086207465, + "gap_ms": [ + 152.8672482818365, + 142.33095478266478 + ], + "text": "Hello, welcome to the voice synthesis benchmark test.", + "warmup": false + }, + { + "request_id": "r0_1", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 27.332807891070843, + "gap_ms": 27.332807891070843, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 182.31825530529022, + "gap_ms": 154.98544741421938, + "samples": 48000 + } + ], + "t_final_ms": 309.03063993901014, + "ttfa_ms": 27.332807891070843, + "playback_start_ms": 102.31825530529022, + "stall_at_ttfa_ms": 74.98544741421938, + "n_chunks": 2, + "streamed_audio_s": 2.08, + "tail_audio_s": 1.84, + "total_wall_s": 0.30903063993901014, + "rtf_streamed": 0.0876530073583126, + "rtf_total": 0.07883434692321688, + "gap_ms": [ + 154.98544741421938 + ], + "text": "She said she would be here by noon, but nobody showed up.", + "warmup": false + }, + { + "request_id": "r0_2", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 33.08893088251352, + "gap_ms": 33.08893088251352, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 178.0979409813881, + "gap_ms": 145.00901009887457, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 325.0236972235143, + "gap_ms": 146.92575624212623, + "samples": 48000 + } + ], + "t_final_ms": 376.1107139289379, + "ttfa_ms": 33.08893088251352, + "playback_start_ms": 98.09794098138809, + "stall_at_ttfa_ms": 65.00901009887457, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.72, + "total_wall_s": 0.3761107139289379, + "rtf_streamed": 0.07966267088811625, + "rtf_total": 0.0783563987351954, + "gap_ms": [ + 145.00901009887457, + 146.92575624212623 + ], + "text": "The quick brown fox jumps over the lazy dog near the riverbank.", + "warmup": false + }, + { + "request_id": "r0_3", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 31.703157350420952, + "gap_ms": 31.703157350420952, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 177.44795512408018, + "gap_ms": 145.74479777365923, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 323.0836079455912, + "gap_ms": 145.63565282151103, + "samples": 48000 + } + ], + "t_final_ms": 365.7913221977651, + "ttfa_ms": 31.703157350420952, + "playback_start_ms": 97.44795512408018, + "stall_at_ttfa_ms": 65.74479777365923, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.64, + "total_wall_s": 0.3657913221977651, + "rtf_streamed": 0.07918715881019392, + "rtf_total": 0.07749816148257736, + "gap_ms": [ + 145.74479777365923, + 145.63565282151103 + ], + "text": "I can't believe how beautiful the sunset looks from up here on the mountain.", + "warmup": false + }, + { + "request_id": "r0_4", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 25.658213067799807, + "gap_ms": 25.658213067799807, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 175.3703560680151, + "gap_ms": 149.7121430002153, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 322.09583604708314, + "gap_ms": 146.72547997906804, + "samples": 48000 + } + ], + "t_final_ms": 438.12064873054624, + "ttfa_ms": 25.658213067799807, + "playback_start_ms": 95.3703560680151, + "stall_at_ttfa_ms": 69.71214300021529, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.68, + "total_wall_s": 0.43812064873054624, + "rtf_streamed": 0.07894505785467724, + "rtf_total": 0.07606261262683095, + "gap_ms": [ + 149.7121430002153, + 146.72547997906804 + ], + "text": "Please remember to bring your identification documents to the appointment tomorrow morning.", + "warmup": false + }, + { + "request_id": "r0_5", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 28.26229529455304, + "gap_ms": 28.26229529455304, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 174.8103192076087, + "gap_ms": 146.54802391305566, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 320.9355133585632, + "gap_ms": 146.12519415095448, + "samples": 48000 + }, + { + "idx": 3, + "t_ms": 463.98471808061004, + "gap_ms": 143.04920472204685, + "samples": 48000 + } + ], + "t_final_ms": 472.7733642794192, + "ttfa_ms": 28.26229529455304, + "playback_start_ms": 94.8103192076087, + "stall_at_ttfa_ms": 66.54802391305566, + "n_chunks": 4, + "streamed_audio_s": 6.08, + "tail_audio_s": 0.08, + "total_wall_s": 0.4727733642794192, + "rtf_streamed": 0.07631327600010034, + "rtf_total": 0.07674892277263298, + "gap_ms": [ + 146.54802391305566, + 146.12519415095448, + 143.04920472204685 + ], + "text": "Have you ever wondered what it would be like to travel through time and visit ancient civilizations?", + "warmup": false + }, + { + "request_id": "r0_6", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 31.987276393920183, + "gap_ms": 31.987276393920183, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 179.85222022980452, + "gap_ms": 147.86494383588433, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 323.06918734684587, + "gap_ms": 143.21696711704135, + "samples": 48000 + } + ], + "t_final_ms": 449.94986103847623, + "ttfa_ms": 31.987276393920183, + "playback_start_ms": 99.85222022980452, + "stall_at_ttfa_ms": 67.86494383588433, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.84, + "total_wall_s": 0.44994986103847623, + "rtf_streamed": 0.07918362434971712, + "rtf_total": 0.0760050440943372, + "gap_ms": [ + 147.86494383588433, + 143.21696711704135 + ], + "text": "The restaurant on the corner serves the best pasta I have ever tasted in my entire life.", + "warmup": false + }, + { + "request_id": "r0_7", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 28.33425998687744, + "gap_ms": 28.33425998687744, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 178.05370362475514, + "gap_ms": 149.7194436378777, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 321.5717510320246, + "gap_ms": 143.51804740726948, + "samples": 48000 + } + ], + "t_final_ms": 386.41078071668744, + "ttfa_ms": 28.33425998687744, + "playback_start_ms": 98.05370362475514, + "stall_at_ttfa_ms": 69.7194436378777, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.96, + "total_wall_s": 0.38641078071668744, + "rtf_streamed": 0.07881660564510408, + "rtf_total": 0.07666880569775544, + "gap_ms": [ + 149.7194436378777, + 143.51804740726948 + ], + "text": "After the meeting, we should discuss the quarterly results and plan for the next phase.", + "warmup": false + }, + { + "request_id": "r0_8", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 29.327966272830963, + "gap_ms": 29.327966272830963, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 176.19290202856064, + "gap_ms": 146.86493575572968, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 322.4714258685708, + "gap_ms": 146.27852384001017, + "samples": 48000 + }, + { + "idx": 3, + "t_ms": 467.18720719218254, + "gap_ms": 144.71578132361174, + "samples": 48000 + } + ], + "t_final_ms": 508.5371150635183, + "ttfa_ms": 29.327966272830963, + "playback_start_ms": 96.19290202856064, + "stall_at_ttfa_ms": 66.86493575572968, + "n_chunks": 4, + "streamed_audio_s": 6.08, + "tail_audio_s": 0.64, + "total_wall_s": 0.5085371150635183, + "rtf_streamed": 0.07684000118292476, + "rtf_total": 0.07567516593207117, + "gap_ms": [ + 146.86493575572968, + 146.27852384001017, + 144.71578132361174 + ], + "text": "Learning a new language takes patience, practice, and a genuine curiosity about other cultures.", + "warmup": false + }, + { + "request_id": "r0_9", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 28.110348153859377, + "gap_ms": 28.110348153859377, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 174.01733715087175, + "gap_ms": 145.90698899701238, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 321.35757291689515, + "gap_ms": 147.3402357660234, + "samples": 48000 + } + ], + "t_final_ms": 398.7006740644574, + "ttfa_ms": 28.110348153859377, + "playback_start_ms": 94.01733715087175, + "stall_at_ttfa_ms": 65.90698899701238, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.12, + "total_wall_s": 0.3987006740644574, + "rtf_streamed": 0.07876411100904293, + "rtf_total": 0.07667320655085719, + "gap_ms": [ + 145.90698899701238, + 147.3402357660234 + ], + "text": "The train leaves at half past seven, so we need to arrive at the station before then.", + "warmup": false + }, + { + "request_id": "r0_10", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 25.723253842443228, + "gap_ms": 25.723253842443228, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 172.1690990962088, + "gap_ms": 146.44584525376558, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 321.79576624184847, + "gap_ms": 149.62666714563966, + "samples": 48000 + } + ], + "t_final_ms": 419.0051881596446, + "ttfa_ms": 25.723253842443228, + "playback_start_ms": 92.16909909620881, + "stall_at_ttfa_ms": 66.44584525376558, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.44, + "total_wall_s": 0.4190051881596446, + "rtf_streamed": 0.07887151133378639, + "rtf_total": 0.07590673698544287, + "gap_ms": [ + 146.44584525376558, + 149.62666714563966 + ], + "text": "Could you please turn down the music a little bit, I'm trying to concentrate on my work.", + "warmup": false + }, + { + "request_id": "r0_11", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 27.757233008742332, + "gap_ms": 27.757233008742332, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 172.31305316090584, + "gap_ms": 144.5558201521635, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 317.1058502048254, + "gap_ms": 144.79279704391956, + "samples": 48000 + } + ], + "t_final_ms": 403.83445331826806, + "ttfa_ms": 27.757233008742332, + "playback_start_ms": 92.31305316090584, + "stall_at_ttfa_ms": 64.5558201521635, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.2, + "total_wall_s": 0.40383445331826806, + "rtf_streamed": 0.07772202210902583, + "rtf_total": 0.0764837979769447, + "gap_ms": [ + 144.5558201521635, + 144.79279704391956 + ], + "text": "It was a dark and stormy night when the old lighthouse keeper heard a knock at the door.", + "warmup": false + }, + { + "request_id": "r1_0", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 26.762599125504494, + "gap_ms": 26.762599125504494, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 174.4638541713357, + "gap_ms": 147.7012550458312, + "samples": 48000 + } + ], + "t_final_ms": 306.3589748926461, + "ttfa_ms": 26.762599125504494, + "playback_start_ms": 94.4638541713357, + "stall_at_ttfa_ms": 67.7012550458312, + "n_chunks": 2, + "streamed_audio_s": 2.08, + "tail_audio_s": 1.92, + "total_wall_s": 0.3063589748926461, + "rtf_streamed": 0.08387685296698831, + "rtf_total": 0.07658974372316152, + "gap_ms": [ + 147.7012550458312 + ], + "text": "Hello, welcome to the voice synthesis benchmark test.", + "warmup": false + }, + { + "request_id": "r1_1", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 32.722072675824165, + "gap_ms": 32.722072675824165, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 170.89650174602866, + "gap_ms": 138.1744290702045, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 318.5585429891944, + "gap_ms": 147.66204124316573, + "samples": 48000 + } + ], + "t_final_ms": 335.86853090673685, + "ttfa_ms": 32.722072675824165, + "playback_start_ms": 90.89650174602866, + "stall_at_ttfa_ms": 58.174429070204496, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.24, + "total_wall_s": 0.33586853090673685, + "rtf_streamed": 0.07807807426205744, + "rtf_total": 0.07774734511730019, + "gap_ms": [ + 138.1744290702045, + 147.66204124316573 + ], + "text": "She said she would be here by noon, but nobody showed up.", + "warmup": false + }, + { + "request_id": "r1_2", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 26.28631517291069, + "gap_ms": 26.28631517291069, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 172.43924411013722, + "gap_ms": 146.15292893722653, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 318.7859049066901, + "gap_ms": 146.3466607965529, + "samples": 48000 + } + ], + "t_final_ms": 353.31569192931056, + "ttfa_ms": 26.28631517291069, + "playback_start_ms": 92.43924411013722, + "stall_at_ttfa_ms": 66.15292893722653, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.48, + "total_wall_s": 0.35331569192931056, + "rtf_streamed": 0.07813380022222798, + "rtf_total": 0.0774815113880067, + "gap_ms": [ + 146.15292893722653, + 146.3466607965529 + ], + "text": "The quick brown fox jumps over the lazy dog near the riverbank.", + "warmup": false + }, + { + "request_id": "r1_3", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 29.720925260335207, + "gap_ms": 29.720925260335207, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 178.22389025241137, + "gap_ms": 148.50296499207616, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 322.42285599932075, + "gap_ms": 144.19896574690938, + "samples": 48000 + } + ], + "t_final_ms": 379.1482551023364, + "ttfa_ms": 29.720925260335207, + "playback_start_ms": 98.22389025241137, + "stall_at_ttfa_ms": 68.50296499207616, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.88, + "total_wall_s": 0.3791482551023364, + "rtf_streamed": 0.07902520980375509, + "rtf_total": 0.07644118046418073, + "gap_ms": [ + 148.50296499207616, + 144.19896574690938 + ], + "text": "I can't believe how beautiful the sunset looks from up here on the mountain.", + "warmup": false + }, + { + "request_id": "r1_4", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 25.721156038343906, + "gap_ms": 25.721156038343906, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 174.36474980786443, + "gap_ms": 148.64359376952052, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 317.03630182892084, + "gap_ms": 142.6715520210564, + "samples": 48000 + } + ], + "t_final_ms": 411.2470089457929, + "ttfa_ms": 25.721156038343906, + "playback_start_ms": 94.36474980786443, + "stall_at_ttfa_ms": 68.64359376952052, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.36, + "total_wall_s": 0.4112470089457929, + "rtf_streamed": 0.07770497593846099, + "rtf_total": 0.07559687664444723, + "gap_ms": [ + 148.64359376952052, + 142.6715520210564 + ], + "text": "Please remember to bring your identification documents to the appointment tomorrow morning.", + "warmup": false + }, + { + "request_id": "r1_5", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 26.86535380780697, + "gap_ms": 26.86535380780697, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 173.88766119256616, + "gap_ms": 147.0223073847592, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 321.2597630918026, + "gap_ms": 147.37210189923644, + "samples": 48000 + } + ], + "t_final_ms": 414.1071098856628, + "ttfa_ms": 26.86535380780697, + "playback_start_ms": 93.88766119256616, + "stall_at_ttfa_ms": 67.02230738475919, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.36, + "total_wall_s": 0.4141071098856628, + "rtf_streamed": 0.07874013801269672, + "rtf_total": 0.076122630493688, + "gap_ms": [ + 147.0223073847592, + 147.37210189923644 + ], + "text": "Have you ever wondered what it would be like to travel through time and visit ancient civilizations?", + "warmup": false + }, + { + "request_id": "r1_6", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 27.762351091951132, + "gap_ms": 27.762351091951132, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 175.07842043414712, + "gap_ms": 147.316069342196, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 317.810888402164, + "gap_ms": 142.73246796801686, + "samples": 48000 + } + ], + "t_final_ms": 416.5652091614902, + "ttfa_ms": 27.762351091951132, + "playback_start_ms": 95.07842043414712, + "stall_at_ttfa_ms": 67.31606934219599, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.36, + "total_wall_s": 0.4165652091614902, + "rtf_streamed": 0.07789482558876568, + "rtf_total": 0.0765744869782151, + "gap_ms": [ + 147.316069342196, + 142.73246796801686 + ], + "text": "The restaurant on the corner serves the best pasta I have ever tasted in my entire life.", + "warmup": false + }, + { + "request_id": "r1_7", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 31.923953909426928, + "gap_ms": 31.923953909426928, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 181.91075883805752, + "gap_ms": 149.9868049286306, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 367.7546880207956, + "gap_ms": 185.84392918273807, + "samples": 48000 + } + ], + "t_final_ms": 479.7564628534019, + "ttfa_ms": 31.923953909426928, + "playback_start_ms": 101.91075883805752, + "stall_at_ttfa_ms": 69.98680492863059, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.6, + "total_wall_s": 0.4797564628534019, + "rtf_streamed": 0.09013595294627343, + "rtf_total": 0.08446416599531724, + "gap_ms": [ + 149.9868049286306, + 185.84392918273807 + ], + "text": "After the meeting, we should discuss the quarterly results and plan for the next phase.", + "warmup": false + }, + { + "request_id": "r1_8", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 27.569978032261133, + "gap_ms": 27.569978032261133, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 179.49419608339667, + "gap_ms": 151.92421805113554, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 320.1808310113847, + "gap_ms": 140.68663492798805, + "samples": 48000 + }, + { + "idx": 3, + "t_ms": 554.2884739115834, + "gap_ms": 234.1076429001987, + "samples": 48000 + } + ], + "t_final_ms": 658.0129712820053, + "ttfa_ms": 27.569978032261133, + "playback_start_ms": 99.49419608339667, + "stall_at_ttfa_ms": 71.92421805113554, + "n_chunks": 4, + "streamed_audio_s": 6.08, + "tail_audio_s": 1.44, + "total_wall_s": 0.6580129712820053, + "rtf_streamed": 0.09116586741966833, + "rtf_total": 0.08750172490452199, + "gap_ms": [ + 151.92421805113554, + 140.68663492798805, + 234.1076429001987 + ], + "text": "Learning a new language takes patience, practice, and a genuine curiosity about other cultures.", + "warmup": false + }, + { + "request_id": "r1_9", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 36.732533015310764, + "gap_ms": 36.732533015310764, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 180.19061721861362, + "gap_ms": 143.45808420330286, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 324.2692584171891, + "gap_ms": 144.0786411985755, + "samples": 48000 + } + ], + "t_final_ms": 448.08512832969427, + "ttfa_ms": 36.732533015310764, + "playback_start_ms": 100.19061721861362, + "stall_at_ttfa_ms": 63.45808420330286, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.76, + "total_wall_s": 0.44808512832969427, + "rtf_streamed": 0.07947775941597772, + "rtf_total": 0.07672690553590655, + "gap_ms": [ + 143.45808420330286, + 144.0786411985755 + ], + "text": "The train leaves at half past seven, so we need to arrive at the station before then.", + "warmup": false + }, + { + "request_id": "r1_10", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 26.32530825212598, + "gap_ms": 26.32530825212598, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 173.06610522791743, + "gap_ms": 146.74079697579145, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 317.1142120845616, + "gap_ms": 144.04810685664415, + "samples": 48000 + } + ], + "t_final_ms": 422.7745649404824, + "ttfa_ms": 26.32530825212598, + "playback_start_ms": 93.06610522791743, + "stall_at_ttfa_ms": 66.74079697579145, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.52, + "total_wall_s": 0.4227745649404824, + "rtf_streamed": 0.07772407158935332, + "rtf_total": 0.07549545802508614, + "gap_ms": [ + 146.74079697579145, + 144.04810685664415 + ], + "text": "Could you please turn down the music a little bit, I'm trying to concentrate on my work.", + "warmup": false + }, + { + "request_id": "r1_11", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 27.94094802811742, + "gap_ms": 27.94094802811742, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 176.63592705503106, + "gap_ms": 148.69497902691364, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 320.0336117297411, + "gap_ms": 143.39768467471004, + "samples": 48000 + } + ], + "t_final_ms": 397.6024268195033, + "ttfa_ms": 27.94094802811742, + "playback_start_ms": 96.63592705503106, + "stall_at_ttfa_ms": 68.69497902691364, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.12, + "total_wall_s": 0.3976024268195033, + "rtf_streamed": 0.0784396107180738, + "rtf_total": 0.07646200515759678, + "gap_ms": [ + 148.69497902691364, + 143.39768467471004 + ], + "text": "It was a dark and stormy night when the old lighthouse keeper heard a knock at the door.", + "warmup": false + }, + { + "request_id": "r2_0", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 30.968028120696545, + "gap_ms": 30.968028120696545, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 173.77827176824212, + "gap_ms": 142.81024364754558, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 318.0728107690811, + "gap_ms": 144.294539000839, + "samples": 48000 + } + ], + "t_final_ms": 362.98405611887574, + "ttfa_ms": 30.968028120696545, + "playback_start_ms": 93.77827176824212, + "stall_at_ttfa_ms": 62.810243647545576, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.64, + "total_wall_s": 0.36298405611887574, + "rtf_streamed": 0.0779590222473238, + "rtf_total": 0.0769034017201008, + "gap_ms": [ + 142.81024364754558, + 144.294539000839 + ], + "text": "Hello, welcome to the voice synthesis benchmark test.", + "warmup": false + }, + { + "request_id": "r2_1", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 31.6069140098989, + "gap_ms": 31.6069140098989, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 178.29942516982555, + "gap_ms": 146.69251115992665, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 328.62059818580747, + "gap_ms": 150.3211730159819, + "samples": 48000 + } + ], + "t_final_ms": 370.06718199700117, + "ttfa_ms": 31.6069140098989, + "playback_start_ms": 98.29942516982555, + "stall_at_ttfa_ms": 66.69251115992665, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.64, + "total_wall_s": 0.37006718199700117, + "rtf_streamed": 0.08054426426122732, + "rtf_total": 0.0784040639824155, + "gap_ms": [ + 146.69251115992665, + 150.3211730159819 + ], + "text": "She said she would be here by noon, but nobody showed up.", + "warmup": false + }, + { + "request_id": "r2_2", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 27.569381054490805, + "gap_ms": 27.569381054490805, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 177.85634007304907, + "gap_ms": 150.28695901855826, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 320.0817028991878, + "gap_ms": 142.22536282613873, + "samples": 48000 + } + ], + "t_final_ms": 333.08024471625686, + "ttfa_ms": 27.569381054490805, + "playback_start_ms": 97.85634007304907, + "stall_at_ttfa_ms": 70.28695901855826, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.16, + "total_wall_s": 0.33308024471625686, + "rtf_streamed": 0.07845139776940878, + "rtf_total": 0.07855666148968321, + "gap_ms": [ + 150.28695901855826, + 142.22536282613873 + ], + "text": "The quick brown fox jumps over the lazy dog near the riverbank.", + "warmup": false + }, + { + "request_id": "r2_3", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 26.80440992116928, + "gap_ms": 26.80440992116928, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 174.8776799067855, + "gap_ms": 148.0732699856162, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 318.6982679180801, + "gap_ms": 143.8205880112946, + "samples": 48000 + } + ], + "t_final_ms": 383.4773520939052, + "ttfa_ms": 26.80440992116928, + "playback_start_ms": 94.87767990678549, + "stall_at_ttfa_ms": 68.07326998561621, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.96, + "total_wall_s": 0.3834773520939052, + "rtf_streamed": 0.07811232056815688, + "rtf_total": 0.07608677620910817, + "gap_ms": [ + 148.0732699856162, + 143.8205880112946 + ], + "text": "I can't believe how beautiful the sunset looks from up here on the mountain.", + "warmup": false + }, + { + "request_id": "r2_4", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 27.34302869066596, + "gap_ms": 27.34302869066596, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 175.0699318945408, + "gap_ms": 147.72690320387483, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 318.6228331178427, + "gap_ms": 143.5529012233019, + "samples": 48000 + } + ], + "t_final_ms": 386.191647965461, + "ttfa_ms": 27.34302869066596, + "playback_start_ms": 95.06993189454079, + "stall_at_ttfa_ms": 67.72690320387483, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.96, + "total_wall_s": 0.386191647965461, + "rtf_streamed": 0.07809383164653007, + "rtf_total": 0.076625326977274, + "gap_ms": [ + 147.72690320387483, + 143.5529012233019 + ], + "text": "Please remember to bring your identification documents to the appointment tomorrow morning.", + "warmup": false + }, + { + "request_id": "r2_5", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 26.43038611859083, + "gap_ms": 26.43038611859083, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 177.02200775966048, + "gap_ms": 150.59162164106965, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 320.0798947364092, + "gap_ms": 143.0578869767487, + "samples": 48000 + } + ], + "t_final_ms": 430.67935574799776, + "ttfa_ms": 26.43038611859083, + "playback_start_ms": 97.02200775966048, + "stall_at_ttfa_ms": 70.59162164106965, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.6, + "total_wall_s": 0.43067935574799776, + "rtf_streamed": 0.07845095459225715, + "rtf_total": 0.07582383023732356, + "gap_ms": [ + 150.59162164106965, + 143.0578869767487 + ], + "text": "Have you ever wondered what it would be like to travel through time and visit ancient civilizations?", + "warmup": false + }, + { + "request_id": "r2_6", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 28.798977844417095, + "gap_ms": 28.798977844417095, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 175.47547072172165, + "gap_ms": 146.67649287730455, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 319.32116486132145, + "gap_ms": 143.8456941395998, + "samples": 48000 + } + ], + "t_final_ms": 452.1001628600061, + "ttfa_ms": 28.798977844417095, + "playback_start_ms": 95.47547072172165, + "stall_at_ttfa_ms": 66.67649287730455, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.92, + "total_wall_s": 0.4521001628600061, + "rtf_streamed": 0.07826499138757878, + "rtf_total": 0.07535002714333434, + "gap_ms": [ + 146.67649287730455, + 143.8456941395998 + ], + "text": "The restaurant on the corner serves the best pasta I have ever tasted in my entire life.", + "warmup": false + }, + { + "request_id": "r2_7", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 25.201807264238596, + "gap_ms": 25.201807264238596, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 173.19971602410078, + "gap_ms": 147.99790875986218, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 318.5948911122978, + "gap_ms": 145.395175088197, + "samples": 48000 + } + ], + "t_final_ms": 451.22405141592026, + "ttfa_ms": 25.201807264238596, + "playback_start_ms": 93.19971602410078, + "stall_at_ttfa_ms": 67.99790875986218, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.92, + "total_wall_s": 0.45122405141592026, + "rtf_streamed": 0.07808698311575926, + "rtf_total": 0.07520400856932004, + "gap_ms": [ + 147.99790875986218, + 145.395175088197 + ], + "text": "After the meeting, we should discuss the quarterly results and plan for the next phase.", + "warmup": false + }, + { + "request_id": "r2_8", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 26.75019996240735, + "gap_ms": 26.75019996240735, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 168.24110690504313, + "gap_ms": 141.49090694263577, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 318.2230917736888, + "gap_ms": 149.98198486864567, + "samples": 48000 + }, + { + "idx": 3, + "t_ms": 462.0327455922961, + "gap_ms": 143.80965381860733, + "samples": 48000 + } + ], + "t_final_ms": 486.9955778121948, + "ttfa_ms": 26.75019996240735, + "playback_start_ms": 88.24110690504313, + "stall_at_ttfa_ms": 61.490906942635775, + "n_chunks": 4, + "streamed_audio_s": 6.08, + "tail_audio_s": 0.4, + "total_wall_s": 0.4869955778121948, + "rtf_streamed": 0.07599222789346975, + "rtf_total": 0.07515363855126463, + "gap_ms": [ + 141.49090694263577, + 149.98198486864567, + 143.80965381860733 + ], + "text": "Learning a new language takes patience, practice, and a genuine curiosity about other cultures.", + "warmup": false + }, + { + "request_id": "r2_9", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 30.30406916514039, + "gap_ms": 30.30406916514039, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 178.24034811928868, + "gap_ms": 147.9362789541483, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 322.2105400636792, + "gap_ms": 143.97019194439054, + "samples": 48000 + } + ], + "t_final_ms": 450.34499699249864, + "ttfa_ms": 30.30406916514039, + "playback_start_ms": 98.24034811928868, + "stall_at_ttfa_ms": 67.93627895414829, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.84, + "total_wall_s": 0.45034499699249864, + "rtf_streamed": 0.0789731715842351, + "rtf_total": 0.07607179003251666, + "gap_ms": [ + 147.9362789541483, + 143.97019194439054 + ], + "text": "The train leaves at half past seven, so we need to arrive at the station before then.", + "warmup": false + }, + { + "request_id": "r2_10", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 29.1077122092247, + "gap_ms": 29.1077122092247, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 175.7540963590145, + "gap_ms": 146.6463841497898, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 319.7092260234058, + "gap_ms": 143.95512966439128, + "samples": 48000 + } + ], + "t_final_ms": 404.04283720999956, + "ttfa_ms": 29.1077122092247, + "playback_start_ms": 95.75409635901451, + "stall_at_ttfa_ms": 66.64638414978981, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.2, + "total_wall_s": 0.40404283720999956, + "rtf_streamed": 0.07836010441750142, + "rtf_total": 0.07652326462310598, + "gap_ms": [ + 146.6463841497898, + 143.95512966439128 + ], + "text": "Could you please turn down the music a little bit, I'm trying to concentrate on my work.", + "warmup": false + }, + { + "request_id": "r2_11", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 26.88279887661338, + "gap_ms": 26.88279887661338, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 174.60785107687116, + "gap_ms": 147.72505220025778, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 320.8113219588995, + "gap_ms": 146.20347088202834, + "samples": 48000 + } + ], + "t_final_ms": 388.7102617882192, + "ttfa_ms": 26.88279887661338, + "playback_start_ms": 94.60785107687116, + "stall_at_ttfa_ms": 67.72505220025778, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.96, + "total_wall_s": 0.3887102617882192, + "rtf_streamed": 0.0786302259703185, + "rtf_total": 0.07712505194210699, + "gap_ms": [ + 147.72505220025778, + 146.20347088202834 + ], + "text": "It was a dark and stormy night when the old lighthouse keeper heard a knock at the door.", + "warmup": false + } + ], + "summary": { + "n_requests": 36, + "ttfa_ms": { + "n": 36, + "p50": 27.757233008742332, + "p90": 31.987276393920183, + "max": 36.732533015310764, + "mean": 28.557164607466095 + }, + "playback_start_ms": { + "n": 36, + "p50": 95.3703560680151, + "p90": 99.85222022980452, + "max": 102.31825530529022, + "mean": 95.92376495142364 + }, + "stall_at_ttfa_ms": { + "n": 36, + "p50": 67.31606934219599, + "p90": 70.59162164106965, + "max": 74.98544741421938, + "mean": 67.36660034395754 + }, + "rtf_streamed": { + "n": 36, + "p50": 0.07845139776940878, + "p90": 0.08387685296698831, + "max": 0.09116586741966833, + "mean": 0.07947681071028984 + }, + "rtf_total": { + "n": 36, + "p50": 0.0765744869782151, + "p90": 0.07855666148968321, + "max": 0.08750172490452199, + "mean": 0.07715958289732197 + }, + "gap_ms": { + "n": 74, + "p50": 146.3466607965529, + "p90": 150.28695901855826, + "max": 234.1076429001987, + "mean": 147.8791764270313 + }, + "total_audio_s": 150.88, + "total_wall_s": 14.805772406514734 + } +} \ No newline at end of file diff --git a/benchmarks/tts/edge_results/stream_latency/Qwen3-TTS-12Hz-1.7B-CustomVoice/default/20260909-103504_baseline.json b/benchmarks/tts/edge_results/stream_latency/Qwen3-TTS-12Hz-1.7B-CustomVoice/default/20260909-103504_baseline.json new file mode 100644 index 000000000..81ece92c4 --- /dev/null +++ b/benchmarks/tts/edge_results/stream_latency/Qwen3-TTS-12Hz-1.7B-CustomVoice/default/20260909-103504_baseline.json @@ -0,0 +1,1620 @@ +{ + "schema": 1, + "meta": { + "model": "Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice", + "deploy_config": null, + "query_type": "CustomVoice", + "txt_prompts": "benchmarks/tts/prompts_12.txt", + "git_sha": "b22dcc59", + "gpu_name": "NVIDIA L20X", + "cuda_visible_devices": "7", + "timestamp": "20260909-103504", + "argv": [ + "--query-type", + "CustomVoice", + "--txt-prompts", + "benchmarks/tts/prompts_12.txt", + "--init-timeout", + "1500", + "--stage-init-timeout", + "900" + ], + "bench_args": { + "model": "Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice", + "warmup": 1, + "repeat": 3, + "init_only": false, + "tag": "baseline", + "output_dir": "/data/zhoutaichang/embedding_infer/vllm-omni/benchmarks/tts/edge_results/stream_latency", + "save_wav": false, + "gpu_poll_interval_s": 0.2, + "parallel_stage_init": false, + "deploy_profile": null, + "bench_help": false + }, + "warmup": 1, + "repeat": 3, + "init_only": false + }, + "init": { + "init_s": 271.4249983429909, + "init_timeline": null + }, + "resources": { + "peak_rss_bytes": 9100095488, + "peak_gpu_mem_mib": 46272, + "gpu_mem_samples": 312, + "gpu_mem_mode": "per-process" + }, + "requests": [ + { + "request_id": "w0", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 516.142345033586, + "gap_ms": 516.142345033586, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 682.061015162617, + "gap_ms": 165.91867012903094, + "samples": 48000 + } + ], + "t_final_ms": 830.2918029949069, + "ttfa_ms": 516.142345033586, + "playback_start_ms": 602.061015162617, + "stall_at_ttfa_ms": 85.91867012903094, + "n_chunks": 2, + "streamed_audio_s": 2.08, + "tail_audio_s": 1.92, + "total_wall_s": 0.8302918029949069, + "rtf_streamed": 0.327913949597412, + "rtf_total": 0.20757295074872673, + "gap_ms": [ + 165.91867012903094 + ], + "text": "Hello, welcome to the voice synthesis benchmark test.", + "warmup": true + }, + { + "request_id": "r0_0", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 29.36774305999279, + "gap_ms": 29.36774305999279, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 192.99644930288196, + "gap_ms": 163.62870624288917, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 354.2883060872555, + "gap_ms": 161.29185678437352, + "samples": 48000 + } + ], + "t_final_ms": 362.16922709718347, + "ttfa_ms": 29.36774305999279, + "playback_start_ms": 112.99644930288196, + "stall_at_ttfa_ms": 83.62870624288917, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.08, + "total_wall_s": 0.36216922709718347, + "rtf_streamed": 0.0868353691390332, + "rtf_total": 0.08705991035989986, + "gap_ms": [ + 163.62870624288917, + 161.29185678437352 + ], + "text": "Hello, welcome to the voice synthesis benchmark test.", + "warmup": false + }, + { + "request_id": "r0_1", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 32.90026495233178, + "gap_ms": 32.90026495233178, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 200.3909139893949, + "gap_ms": 167.49064903706312, + "samples": 48000 + } + ], + "t_final_ms": 312.4939911067486, + "ttfa_ms": 32.90026495233178, + "playback_start_ms": 120.3909139893949, + "stall_at_ttfa_ms": 87.49064903706312, + "n_chunks": 2, + "streamed_audio_s": 2.08, + "tail_audio_s": 1.44, + "total_wall_s": 0.3124939911067486, + "rtf_streamed": 0.09634178557182448, + "rtf_total": 0.08877670201896266, + "gap_ms": [ + 167.49064903706312 + ], + "text": "She said she would be here by noon, but nobody showed up.", + "warmup": false + }, + { + "request_id": "r0_2", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 32.138410955667496, + "gap_ms": 32.138410955667496, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 193.920633289963, + "gap_ms": 161.7822223342955, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 530.4645663127303, + "gap_ms": 336.5439330227673, + "samples": 48000 + } + ], + "t_final_ms": 536.0107142478228, + "ttfa_ms": 32.138410955667496, + "playback_start_ms": 113.920633289963, + "stall_at_ttfa_ms": 81.78222233429551, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.32, + "total_wall_s": 0.5360107142478228, + "rtf_streamed": 0.13001582507664958, + "rtf_total": 0.12182061687450517, + "gap_ms": [ + 161.7822223342955, + 336.5439330227673 + ], + "text": "The quick brown fox jumps over the lazy dog near the riverbank.", + "warmup": false + }, + { + "request_id": "r0_3", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 36.25373309478164, + "gap_ms": 36.25373309478164, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 195.77294727787375, + "gap_ms": 159.51921418309212, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 359.1875550337136, + "gap_ms": 163.41460775583982, + "samples": 48000 + } + ], + "t_final_ms": 489.19968912377954, + "ttfa_ms": 36.25373309478164, + "playback_start_ms": 115.77294727787375, + "stall_at_ttfa_ms": 79.51921418309212, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.68, + "total_wall_s": 0.48919968912377954, + "rtf_streamed": 0.0880361654494396, + "rtf_total": 0.0849305015839895, + "gap_ms": [ + 159.51921418309212, + 163.41460775583982 + ], + "text": "I can't believe how beautiful the sunset looks from up here on the mountain.", + "warmup": false + }, + { + "request_id": "r0_4", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 29.801278840750456, + "gap_ms": 29.801278840750456, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 199.5363961905241, + "gap_ms": 169.73511734977365, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 362.2526889666915, + "gap_ms": 162.7162927761674, + "samples": 48000 + }, + { + "idx": 3, + "t_ms": 522.0494838431478, + "gap_ms": 159.79679487645626, + "samples": 48000 + } + ], + "t_final_ms": 535.0989061407745, + "ttfa_ms": 29.801278840750456, + "playback_start_ms": 119.5363961905241, + "stall_at_ttfa_ms": 89.73511734977365, + "n_chunks": 4, + "streamed_audio_s": 6.08, + "tail_audio_s": 0.16, + "total_wall_s": 0.5350989061407745, + "rtf_streamed": 0.08586340194788614, + "rtf_total": 0.08575302983025232, + "gap_ms": [ + 169.73511734977365, + 162.7162927761674, + 159.79679487645626 + ], + "text": "Please remember to bring your identification documents to the appointment tomorrow morning.", + "warmup": false + }, + { + "request_id": "r0_5", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 29.629172291606665, + "gap_ms": 29.629172291606665, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 192.9285591468215, + "gap_ms": 163.29938685521483, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 356.3054339028895, + "gap_ms": 163.376874756068, + "samples": 48000 + } + ], + "t_final_ms": 470.78959411010146, + "ttfa_ms": 29.629172291606665, + "playback_start_ms": 112.9285591468215, + "stall_at_ttfa_ms": 83.29938685521483, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.44, + "total_wall_s": 0.47078959411010146, + "rtf_streamed": 0.08732976321149252, + "rtf_total": 0.08528796994748215, + "gap_ms": [ + 163.29938685521483, + 163.376874756068 + ], + "text": "Have you ever wondered what it would be like to travel through time and visit ancient civilizations?", + "warmup": false + }, + { + "request_id": "r0_6", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 33.94488291814923, + "gap_ms": 33.94488291814923, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 199.08129563555121, + "gap_ms": 165.13641271740198, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 359.10550272092223, + "gap_ms": 160.02420708537102, + "samples": 48000 + }, + { + "idx": 3, + "t_ms": 519.6888619102538, + "gap_ms": 160.58335918933153, + "samples": 48000 + } + ], + "t_final_ms": 609.46213779971, + "ttfa_ms": 33.94488291814923, + "playback_start_ms": 119.08129563555121, + "stall_at_ttfa_ms": 85.13641271740198, + "n_chunks": 4, + "streamed_audio_s": 6.08, + "tail_audio_s": 1.12, + "total_wall_s": 0.60946213779971, + "rtf_streamed": 0.0854751417615549, + "rtf_total": 0.08464751913884862, + "gap_ms": [ + 165.13641271740198, + 160.02420708537102, + 160.58335918933153 + ], + "text": "The restaurant on the corner serves the best pasta I have ever tasted in my entire life.", + "warmup": false + }, + { + "request_id": "r0_7", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 33.49698521196842, + "gap_ms": 33.49698521196842, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 197.76567909866571, + "gap_ms": 164.2686938866973, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 359.96605595573783, + "gap_ms": 162.20037685707211, + "samples": 48000 + } + ], + "t_final_ms": 439.7185770794749, + "ttfa_ms": 33.49698521196842, + "playback_start_ms": 117.76567909866571, + "stall_at_ttfa_ms": 84.26869388669729, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.04, + "total_wall_s": 0.4397185770794749, + "rtf_streamed": 0.08822697449895535, + "rtf_total": 0.08588253458583495, + "gap_ms": [ + 164.2686938866973, + 162.20037685707211 + ], + "text": "After the meeting, we should discuss the quarterly results and plan for the next phase.", + "warmup": false + }, + { + "request_id": "r0_8", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 31.103891786187887, + "gap_ms": 31.103891786187887, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 199.10128274932504, + "gap_ms": 167.99739096313715, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 363.93954372033477, + "gap_ms": 164.83826097100973, + "samples": 48000 + }, + { + "idx": 3, + "t_ms": 532.6215806417167, + "gap_ms": 168.68203692138195, + "samples": 48000 + } + ], + "t_final_ms": 663.2288065738976, + "ttfa_ms": 31.103891786187887, + "playback_start_ms": 119.10128274932504, + "stall_at_ttfa_ms": 87.99739096313715, + "n_chunks": 4, + "streamed_audio_s": 6.08, + "tail_audio_s": 1.68, + "total_wall_s": 0.6632288065738976, + "rtf_streamed": 0.0876022336581771, + "rtf_total": 0.08546762971313114, + "gap_ms": [ + 167.99739096313715, + 164.83826097100973, + 168.68203692138195 + ], + "text": "Learning a new language takes patience, practice, and a genuine curiosity about other cultures.", + "warmup": false + }, + { + "request_id": "r0_9", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 33.16293004900217, + "gap_ms": 33.16293004900217, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 195.05384098738432, + "gap_ms": 161.89091093838215, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 358.01666090264916, + "gap_ms": 162.96281991526484, + "samples": 48000 + } + ], + "t_final_ms": 488.32017416134477, + "ttfa_ms": 33.16293004900217, + "playback_start_ms": 115.05384098738432, + "stall_at_ttfa_ms": 81.89091093838215, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.68, + "total_wall_s": 0.48832017416134477, + "rtf_streamed": 0.08774918159378656, + "rtf_total": 0.08477780801412235, + "gap_ms": [ + 161.89091093838215, + 162.96281991526484 + ], + "text": "The train leaves at half past seven, so we need to arrive at the station before then.", + "warmup": false + }, + { + "request_id": "r0_10", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 33.247143030166626, + "gap_ms": 33.247143030166626, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 198.6272782087326, + "gap_ms": 165.38013517856598, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 366.18478083983064, + "gap_ms": 167.55750263109803, + "samples": 48000 + } + ], + "t_final_ms": 446.4674908667803, + "ttfa_ms": 33.247143030166626, + "playback_start_ms": 118.6272782087326, + "stall_at_ttfa_ms": 85.38013517856598, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.12, + "total_wall_s": 0.4464674908667803, + "rtf_streamed": 0.08975117177446829, + "rtf_total": 0.0858591328589962, + "gap_ms": [ + 165.38013517856598, + 167.55750263109803 + ], + "text": "Could you please turn down the music a little bit, I'm trying to concentrate on my work.", + "warmup": false + }, + { + "request_id": "r0_11", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 33.07389514520764, + "gap_ms": 33.07389514520764, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 196.00714975968003, + "gap_ms": 162.9332546144724, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 360.09372491389513, + "gap_ms": 164.0865751542151, + "samples": 48000 + } + ], + "t_final_ms": 402.1885311231017, + "ttfa_ms": 33.07389514520764, + "playback_start_ms": 116.00714975968003, + "stall_at_ttfa_ms": 82.93325461447239, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.56, + "total_wall_s": 0.4021885311231017, + "rtf_streamed": 0.08825826591026842, + "rtf_total": 0.08667856274204777, + "gap_ms": [ + 162.9332546144724, + 164.0865751542151 + ], + "text": "It was a dark and stormy night when the old lighthouse keeper heard a knock at the door.", + "warmup": false + }, + { + "request_id": "r1_0", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 34.43072270601988, + "gap_ms": 34.43072270601988, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 199.15949273854494, + "gap_ms": 164.72877003252506, + "samples": 48000 + } + ], + "t_final_ms": 322.3587707616389, + "ttfa_ms": 34.43072270601988, + "playback_start_ms": 119.15949273854494, + "stall_at_ttfa_ms": 84.72877003252506, + "n_chunks": 2, + "streamed_audio_s": 2.08, + "tail_audio_s": 1.6, + "total_wall_s": 0.3223587707616389, + "rtf_streamed": 0.09574975612430045, + "rtf_total": 0.08759749205479317, + "gap_ms": [ + 164.72877003252506 + ], + "text": "Hello, welcome to the voice synthesis benchmark test.", + "warmup": false + }, + { + "request_id": "r1_1", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 32.67227066680789, + "gap_ms": 32.67227066680789, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 197.7998809888959, + "gap_ms": 165.127610322088, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 357.5614229775965, + "gap_ms": 159.76154198870063, + "samples": 48000 + } + ], + "t_final_ms": 365.356033667922, + "ttfa_ms": 32.67227066680789, + "playback_start_ms": 117.7998809888959, + "stall_at_ttfa_ms": 85.127610322088, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.08, + "total_wall_s": 0.365356033667922, + "rtf_streamed": 0.08763760367097954, + "rtf_total": 0.08782596963171202, + "gap_ms": [ + 165.127610322088, + 159.76154198870063 + ], + "text": "She said she would be here by noon, but nobody showed up.", + "warmup": false + }, + { + "request_id": "r1_2", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 31.42575081437826, + "gap_ms": 31.42575081437826, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 199.52413300052285, + "gap_ms": 168.0983821861446, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 358.51368587464094, + "gap_ms": 158.9895528741181, + "samples": 48000 + } + ], + "t_final_ms": 367.32427263632417, + "ttfa_ms": 31.42575081437826, + "playback_start_ms": 119.52413300052285, + "stall_at_ttfa_ms": 88.09838218614459, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.08, + "total_wall_s": 0.36732427263632417, + "rtf_streamed": 0.08787100143986297, + "rtf_total": 0.08829910399911638, + "gap_ms": [ + 168.0983821861446, + 158.9895528741181 + ], + "text": "The quick brown fox jumps over the lazy dog near the riverbank.", + "warmup": false + }, + { + "request_id": "r1_3", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 33.06374419480562, + "gap_ms": 33.06374419480562, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 199.59677197039127, + "gap_ms": 166.53302777558565, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 362.75473423302174, + "gap_ms": 163.15796226263046, + "samples": 48000 + } + ], + "t_final_ms": 410.153501201421, + "ttfa_ms": 33.06374419480562, + "playback_start_ms": 119.59677197039127, + "stall_at_ttfa_ms": 86.53302777558565, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.64, + "total_wall_s": 0.410153501201421, + "rtf_streamed": 0.08891047407672101, + "rtf_total": 0.08689692822064005, + "gap_ms": [ + 166.53302777558565, + 163.15796226263046 + ], + "text": "I can't believe how beautiful the sunset looks from up here on the mountain.", + "warmup": false + }, + { + "request_id": "r1_4", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 27.629636228084564, + "gap_ms": 27.629636228084564, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 192.31106201186776, + "gap_ms": 164.6814257837832, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 354.3332782573998, + "gap_ms": 162.02221624553204, + "samples": 48000 + }, + { + "idx": 3, + "t_ms": 514.7182079963386, + "gap_ms": 160.3849297389388, + "samples": 48000 + } + ], + "t_final_ms": 522.5414554588497, + "ttfa_ms": 27.629636228084564, + "playback_start_ms": 112.31106201186776, + "stall_at_ttfa_ms": 84.6814257837832, + "n_chunks": 4, + "streamed_audio_s": 6.08, + "tail_audio_s": 0.08, + "total_wall_s": 0.5225414554588497, + "rtf_streamed": 0.0846575999993978, + "rtf_total": 0.08482815835370935, + "gap_ms": [ + 164.6814257837832, + 162.02221624553204, + 160.3849297389388 + ], + "text": "Please remember to bring your identification documents to the appointment tomorrow morning.", + "warmup": false + }, + { + "request_id": "r1_5", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 32.07993786782026, + "gap_ms": 32.07993786782026, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 197.3424549214542, + "gap_ms": 165.26251705363393, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 358.4121847525239, + "gap_ms": 161.0697298310697, + "samples": 48000 + } + ], + "t_final_ms": 494.8120419867337, + "ttfa_ms": 32.07993786782026, + "playback_start_ms": 117.34245492145419, + "stall_at_ttfa_ms": 85.26251705363393, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.76, + "total_wall_s": 0.4948120419867337, + "rtf_streamed": 0.08784612371385389, + "rtf_total": 0.08472808938129002, + "gap_ms": [ + 165.26251705363393, + 161.0697298310697 + ], + "text": "Have you ever wondered what it would be like to travel through time and visit ancient civilizations?", + "warmup": false + }, + { + "request_id": "r1_6", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 31.696080695837736, + "gap_ms": 31.696080695837736, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 197.0584187656641, + "gap_ms": 165.36233806982636, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 359.20535307377577, + "gap_ms": 162.14693430811167, + "samples": 48000 + }, + { + "idx": 3, + "t_ms": 516.9564899988472, + "gap_ms": 157.75113692507148, + "samples": 48000 + } + ], + "t_final_ms": 574.9200587160885, + "ttfa_ms": 31.696080695837736, + "playback_start_ms": 117.0584187656641, + "stall_at_ttfa_ms": 85.36233806982636, + "n_chunks": 4, + "streamed_audio_s": 6.08, + "tail_audio_s": 0.72, + "total_wall_s": 0.5749200587160885, + "rtf_streamed": 0.08502573848665251, + "rtf_total": 0.08454706745824832, + "gap_ms": [ + 165.36233806982636, + 162.14693430811167, + 157.75113692507148 + ], + "text": "The restaurant on the corner serves the best pasta I have ever tasted in my entire life.", + "warmup": false + }, + { + "request_id": "r1_7", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 32.05796703696251, + "gap_ms": 32.05796703696251, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 193.25875164940953, + "gap_ms": 161.20078461244702, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 356.81579168885946, + "gap_ms": 163.55704003944993, + "samples": 48000 + } + ], + "t_final_ms": 485.35003885626793, + "ttfa_ms": 32.05796703696251, + "playback_start_ms": 113.25875164940953, + "stall_at_ttfa_ms": 81.20078461244702, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.6, + "total_wall_s": 0.48535003885626793, + "rtf_streamed": 0.08745485090413221, + "rtf_total": 0.08544895050286408, + "gap_ms": [ + 161.20078461244702, + 163.55704003944993 + ], + "text": "After the meeting, we should discuss the quarterly results and plan for the next phase.", + "warmup": false + }, + { + "request_id": "r1_8", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 33.19976478815079, + "gap_ms": 33.19976478815079, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 196.53633376583457, + "gap_ms": 163.33656897768378, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 356.8715429864824, + "gap_ms": 160.3352092206478, + "samples": 48000 + }, + { + "idx": 3, + "t_ms": 520.2861316502094, + "gap_ms": 163.41458866372705, + "samples": 48000 + } + ], + "t_final_ms": 614.2430030740798, + "ttfa_ms": 33.19976478815079, + "playback_start_ms": 116.53633376583457, + "stall_at_ttfa_ms": 83.33656897768378, + "n_chunks": 4, + "streamed_audio_s": 6.08, + "tail_audio_s": 1.2, + "total_wall_s": 0.6142430030740798, + "rtf_streamed": 0.08557337691615287, + "rtf_total": 0.08437403888380215, + "gap_ms": [ + 163.33656897768378, + 160.3352092206478, + 163.41458866372705 + ], + "text": "Learning a new language takes patience, practice, and a genuine curiosity about other cultures.", + "warmup": false + }, + { + "request_id": "r1_9", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 31.981505919247866, + "gap_ms": 31.981505919247866, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 195.35403791815042, + "gap_ms": 163.37253199890256, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 356.88522504642606, + "gap_ms": 161.53118712827563, + "samples": 48000 + } + ], + "t_final_ms": 473.92627876251936, + "ttfa_ms": 31.981505919247866, + "playback_start_ms": 115.35403791815042, + "stall_at_ttfa_ms": 83.37253199890256, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.36, + "total_wall_s": 0.47392627876251936, + "rtf_streamed": 0.08747186888392795, + "rtf_total": 0.08711880124311017, + "gap_ms": [ + 163.37253199890256, + 161.53118712827563 + ], + "text": "The train leaves at half past seven, so we need to arrive at the station before then.", + "warmup": false + }, + { + "request_id": "r1_10", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 32.12744789198041, + "gap_ms": 32.12744789198041, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 194.86408401280642, + "gap_ms": 162.736636120826, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 357.09759779274464, + "gap_ms": 162.23351377993822, + "samples": 48000 + }, + { + "idx": 3, + "t_ms": 519.8257029987872, + "gap_ms": 162.72810520604253, + "samples": 48000 + } + ], + "t_final_ms": 528.6091328598559, + "ttfa_ms": 32.12744789198041, + "playback_start_ms": 114.86408401280642, + "stall_at_ttfa_ms": 82.736636120826, + "n_chunks": 4, + "streamed_audio_s": 6.08, + "tail_audio_s": 0.08, + "total_wall_s": 0.5286091328598559, + "rtf_streamed": 0.08549764851953737, + "rtf_total": 0.08581317091880777, + "gap_ms": [ + 162.736636120826, + 162.23351377993822, + 162.72810520604253 + ], + "text": "Could you please turn down the music a little bit, I'm trying to concentrate on my work.", + "warmup": false + }, + { + "request_id": "r1_11", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 30.374767258763313, + "gap_ms": 30.374767258763313, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 197.46247213333845, + "gap_ms": 167.08770487457514, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 361.06348084285855, + "gap_ms": 163.6010087095201, + "samples": 48000 + } + ], + "t_final_ms": 424.5304302312434, + "ttfa_ms": 30.374767258763313, + "playback_start_ms": 117.46247213333845, + "stall_at_ttfa_ms": 87.08770487457514, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.88, + "total_wall_s": 0.42453043023124337, + "rtf_streamed": 0.08849595118697513, + "rtf_total": 0.08559081254662165, + "gap_ms": [ + 167.08770487457514, + 163.6010087095201 + ], + "text": "It was a dark and stormy night when the old lighthouse keeper heard a knock at the door.", + "warmup": false + }, + { + "request_id": "r2_0", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 35.930318757891655, + "gap_ms": 35.930318757891655, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 197.8219347074628, + "gap_ms": 161.89161594957113, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 363.71457297354937, + "gap_ms": 165.89263826608658, + "samples": 48000 + } + ], + "t_final_ms": 380.3105726838112, + "ttfa_ms": 35.930318757891655, + "playback_start_ms": 117.82193470746279, + "stall_at_ttfa_ms": 81.89161594957113, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.24, + "total_wall_s": 0.3803105726838112, + "rtf_streamed": 0.08914572866998759, + "rtf_total": 0.08803485478791925, + "gap_ms": [ + 161.89161594957113, + 165.89263826608658 + ], + "text": "Hello, welcome to the voice synthesis benchmark test.", + "warmup": false + }, + { + "request_id": "r2_1", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 85.07940405979753, + "gap_ms": 85.07940405979753, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 250.97834784537554, + "gap_ms": 165.898943785578, + "samples": 48000 + } + ], + "t_final_ms": 393.96108826622367, + "ttfa_ms": 85.07940405979753, + "playback_start_ms": 170.97834784537554, + "stall_at_ttfa_ms": 85.89894378557801, + "n_chunks": 2, + "streamed_audio_s": 2.08, + "tail_audio_s": 1.84, + "total_wall_s": 0.39396108826622367, + "rtf_streamed": 0.12066266723335362, + "rtf_total": 0.10050027761893461, + "gap_ms": [ + 165.898943785578 + ], + "text": "She said she would be here by noon, but nobody showed up.", + "warmup": false + }, + { + "request_id": "r2_2", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 26.484066620469093, + "gap_ms": 26.484066620469093, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 190.6544486992061, + "gap_ms": 164.17038207873702, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 355.1582139916718, + "gap_ms": 164.5037652924657, + "samples": 48000 + } + ], + "t_final_ms": 377.948225941509, + "ttfa_ms": 26.484066620469093, + "playback_start_ms": 110.65444869920611, + "stall_at_ttfa_ms": 84.17038207873702, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.32, + "total_wall_s": 0.377948225941509, + "rtf_streamed": 0.08704858186070387, + "rtf_total": 0.08589732407761567, + "gap_ms": [ + 164.17038207873702, + 164.5037652924657 + ], + "text": "The quick brown fox jumps over the lazy dog near the riverbank.", + "warmup": false + }, + { + "request_id": "r2_3", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 28.90548389405012, + "gap_ms": 28.90548389405012, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 193.21021903306246, + "gap_ms": 164.30473513901234, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 358.773672953248, + "gap_ms": 165.56345392018557, + "samples": 48000 + } + ], + "t_final_ms": 422.1228458918631, + "ttfa_ms": 28.90548389405012, + "playback_start_ms": 113.21021903306246, + "stall_at_ttfa_ms": 84.30473513901234, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 0.88, + "total_wall_s": 0.4221228458918631, + "rtf_streamed": 0.08793472376305099, + "rtf_total": 0.08510541247819821, + "gap_ms": [ + 164.30473513901234, + 165.56345392018557 + ], + "text": "I can't believe how beautiful the sunset looks from up here on the mountain.", + "warmup": false + }, + { + "request_id": "r2_4", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 28.569420799613, + "gap_ms": 28.569420799613, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 195.1305828988552, + "gap_ms": 166.5611620992422, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 355.0357399508357, + "gap_ms": 159.9051570519805, + "samples": 48000 + } + ], + "t_final_ms": 508.5050407797098, + "ttfa_ms": 28.569420799613, + "playback_start_ms": 115.13058289885521, + "stall_at_ttfa_ms": 86.56116209924221, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 2.0, + "total_wall_s": 0.5085050407797098, + "rtf_streamed": 0.08701856371344012, + "rtf_total": 0.0836356974966628, + "gap_ms": [ + 166.5611620992422, + 159.9051570519805 + ], + "text": "Please remember to bring your identification documents to the appointment tomorrow morning.", + "warmup": false + }, + { + "request_id": "r2_5", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 29.56665400415659, + "gap_ms": 29.56665400415659, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 191.4986902847886, + "gap_ms": 161.93203628063202, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 354.6110880561173, + "gap_ms": 163.1123977713287, + "samples": 48000 + }, + { + "idx": 3, + "t_ms": 516.8511820957065, + "gap_ms": 162.24009403958917, + "samples": 48000 + } + ], + "t_final_ms": 647.6503303274512, + "ttfa_ms": 29.56665400415659, + "playback_start_ms": 111.49869028478861, + "stall_at_ttfa_ms": 81.93203628063202, + "n_chunks": 4, + "streamed_audio_s": 6.08, + "tail_audio_s": 1.68, + "total_wall_s": 0.6476503303274512, + "rtf_streamed": 0.08500841810784646, + "rtf_total": 0.08346009411436227, + "gap_ms": [ + 161.93203628063202, + 163.1123977713287, + 162.24009403958917 + ], + "text": "Have you ever wondered what it would be like to travel through time and visit ancient civilizations?", + "warmup": false + }, + { + "request_id": "r2_6", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 38.12049189582467, + "gap_ms": 38.12049189582467, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 201.5764657407999, + "gap_ms": 163.45597384497523, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 368.515829090029, + "gap_ms": 166.9393633492291, + "samples": 48000 + } + ], + "t_final_ms": 450.0176450237632, + "ttfa_ms": 38.12049189582467, + "playback_start_ms": 121.5764657407999, + "stall_at_ttfa_ms": 83.45597384497523, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.12, + "total_wall_s": 0.4500176450237632, + "rtf_streamed": 0.09032250712990907, + "rtf_total": 0.08654185481226215, + "gap_ms": [ + 163.45597384497523, + 166.9393633492291 + ], + "text": "The restaurant on the corner serves the best pasta I have ever tasted in my entire life.", + "warmup": false + }, + { + "request_id": "r2_7", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 25.98439110442996, + "gap_ms": 25.98439110442996, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 192.12736189365387, + "gap_ms": 166.1429707892239, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 353.678863029927, + "gap_ms": 161.55150113627315, + "samples": 48000 + } + ], + "t_final_ms": 459.7989791072905, + "ttfa_ms": 25.98439110442996, + "playback_start_ms": 112.12736189365387, + "stall_at_ttfa_ms": 86.14297078922391, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.36, + "total_wall_s": 0.4597989791072905, + "rtf_streamed": 0.08668599584066838, + "rtf_total": 0.0845218711594284, + "gap_ms": [ + 166.1429707892239, + 161.55150113627315 + ], + "text": "After the meeting, we should discuss the quarterly results and plan for the next phase.", + "warmup": false + }, + { + "request_id": "r2_8", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 25.73835989460349, + "gap_ms": 25.73835989460349, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 191.13960396498442, + "gap_ms": 165.40124407038093, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 352.2061789408326, + "gap_ms": 161.0665749758482, + "samples": 48000 + }, + { + "idx": 3, + "t_ms": 513.9299631118774, + "gap_ms": 161.72378417104483, + "samples": 48000 + } + ], + "t_final_ms": 579.4818089343607, + "ttfa_ms": 25.73835989460349, + "playback_start_ms": 111.13960396498442, + "stall_at_ttfa_ms": 85.40124407038093, + "n_chunks": 4, + "streamed_audio_s": 6.08, + "tail_audio_s": 0.88, + "total_wall_s": 0.5794818089343607, + "rtf_streamed": 0.08452795445919037, + "rtf_total": 0.08325888059401734, + "gap_ms": [ + 165.40124407038093, + 161.0665749758482, + 161.72378417104483 + ], + "text": "Learning a new language takes patience, practice, and a genuine curiosity about other cultures.", + "warmup": false + }, + { + "request_id": "r2_9", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 30.063524842262268, + "gap_ms": 30.063524842262268, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 194.22471383586526, + "gap_ms": 164.161188993603, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 358.63392893224955, + "gap_ms": 164.4092150963843, + "samples": 48000 + }, + { + "idx": 3, + "t_ms": 517.5395729020238, + "gap_ms": 158.90564396977425, + "samples": 48000 + } + ], + "t_final_ms": 611.0006128437817, + "ttfa_ms": 30.063524842262268, + "playback_start_ms": 114.22471383586526, + "stall_at_ttfa_ms": 84.16118899360299, + "n_chunks": 4, + "streamed_audio_s": 6.08, + "tail_audio_s": 1.2, + "total_wall_s": 0.6110006128437817, + "rtf_streamed": 0.08512164027993813, + "rtf_total": 0.08392865561040957, + "gap_ms": [ + 164.161188993603, + 164.4092150963843, + 158.90564396977425 + ], + "text": "The train leaves at half past seven, so we need to arrive at the station before then.", + "warmup": false + }, + { + "request_id": "r2_10", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 29.717253986746073, + "gap_ms": 29.717253986746073, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 196.24913716688752, + "gap_ms": 166.53188318014145, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 359.34201115742326, + "gap_ms": 163.09287399053574, + "samples": 48000 + } + ], + "t_final_ms": 497.0862311311066, + "ttfa_ms": 29.717253986746073, + "playback_start_ms": 116.24913716688752, + "stall_at_ttfa_ms": 86.53188318014145, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.76, + "total_wall_s": 0.4970862311311066, + "rtf_streamed": 0.08807402234250569, + "rtf_total": 0.08511750533066895, + "gap_ms": [ + 166.53188318014145, + 163.09287399053574 + ], + "text": "Could you please turn down the music a little bit, I'm trying to concentrate on my work.", + "warmup": false + }, + { + "request_id": "r2_11", + "sr": 24000, + "chunks": [ + { + "idx": 0, + "t_ms": 30.56880459189415, + "gap_ms": 30.56880459189415, + "samples": 1920 + }, + { + "idx": 1, + "t_ms": 196.19254395365715, + "gap_ms": 165.623739361763, + "samples": 48000 + }, + { + "idx": 2, + "t_ms": 358.1041255965829, + "gap_ms": 161.91158164292574, + "samples": 48000 + } + ], + "t_final_ms": 439.9677109904587, + "ttfa_ms": 30.56880459189415, + "playback_start_ms": 116.19254395365715, + "stall_at_ttfa_ms": 85.623739361763, + "n_chunks": 3, + "streamed_audio_s": 4.08, + "tail_audio_s": 1.04, + "total_wall_s": 0.4399677109904587, + "rtf_streamed": 0.08777061901877031, + "rtf_total": 0.08593119355282397, + "gap_ms": [ + 165.623739361763, + 161.91158164292574 + ], + "text": "It was a dark and stormy night when the old lighthouse keeper heard a knock at the door.", + "warmup": false + } + ], + "summary": { + "n_requests": 36, + "ttfa_ms": { + "n": 36, + "p50": 31.981505919247866, + "p90": 35.930318757891655, + "max": 85.07940405979753, + "mean": 32.93300282934474 + }, + "playback_start_ms": { + "n": 36, + "p50": 116.19254395365715, + "p90": 119.59677197039127, + "max": 170.97834784537554, + "mean": 117.56262137606326 + }, + "stall_at_ttfa_ms": { + "n": 36, + "p50": 84.6814257837832, + "p90": 87.49064903706312, + "max": 89.73511734977365, + "mean": 84.62961854671852 + }, + "rtf_streamed": { + "n": 36, + "p50": 0.08763760367097954, + "p90": 0.09574975612430045, + "max": 0.13001582507664958, + "mean": 0.08980551933153874 + }, + "rtf_total": { + "n": 36, + "p50": 0.08559081254662165, + "p90": 0.08829910399911638, + "max": 0.12182061687450517, + "mean": 0.08710955895822475 + }, + "gap_ms": { + "n": 79, + "p50": 163.376874756068, + "p90": 167.08770487457514, + "max": 336.5439330227673, + "mean": 165.67398060726214 + }, + "total_audio_s": 160.88, + "total_wall_s": 17.107123949564993 + } +} \ No newline at end of file