tzhouamandClaude Fable 5.1 b70207f57f [Edge][WP0] Add offline streaming-latency benchmark for chunked TTS ...

Adds benchmarks/tts/stream_latency_bench.py (AsyncOmni driver reusing the
Qwen3-TTS example's prompt builder; --model override, warm-up/repeat,
--init-only, per-chunk arrivals, peak RSS and per-process GPU memory) and
benchmarks/tts/stream_latency_metrics.py (TTFA, underrun-free playback start,
stall at TTFA, RTF streamed/total, chunk gaps, percentiles) with CPU unit
tests. Results go under benchmarks/tts/edge_results/stream_latency/.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

2026-09-09 10:24:53 +08:00

..

duplex

[Benchmark][MiniCPM-o] Add Omni-DuplexEval support (#6634)

2026-08-27 16:31:50 +08:00

metrics

[Bugfix] Avoid invalid TPOT metrics (#6696)

2026-08-29 11:56:37 +00:00

patch

[Core] Graduate MiniCPM-o 4.5 and PersonaPlex full-duplex serving out of experimental (#6196)

2026-09-05 08:20:21 +00:00

benchmark_video_response_encoding.py

[Frontend] Encode interleaved video frames in parallel (#6776)

2026-08-31 07:04:27 +08:00

conftest.py

[Benchmark] Universal TTS benchmark: Qwen3-TTS + VoxCPM2 with 3 task types (voice-clone/default/design) (#2835)

2026-04-24 23:35:33 +08:00

test_accuracy_bench_utils.py

[CI][MiniCPM-o] Add MiniCPM-o 4.5 accuracy and performance coverage (#5524)

2026-08-11 14:53:49 +08:00

test_audio_continuity.py

[Bench] Add audio-streaming continuity metric for TTS (#3618)

2026-05-20 05:08:56 +00:00

test_bench_tts_cli.py

[New Model] Support IndexTTS 2.5 (#5957)

2026-08-11 23:00:35 +00:00

test_daily_omni_local_source.py

[Bugfix][MiniCPM-o] Load Daily-Omni QA metadata from the hub cache (#6276)

2026-08-18 10:53:55 +08:00

test_daily_omni_pack_modes.py

[Bugfix][MiniCPM-o] Bind omni-duplex video frames to the unit they close (#6404)

2026-08-28 16:58:20 +08:00

test_diffusion_backends_metrics.py

[Refactor]Rename diffusion benchmark backend to endpoint (#3137)

2026-05-19 11:30:33 +08:00

test_omniinteract.py

[Core] Graduate MiniCPM-o 4.5 and PersonaPlex full-duplex serving out of experimental (#6196)

2026-09-05 08:20:21 +00:00

test_seed_tts_dataset_variants.py

[Bugfix] Use Seed-TTS reference audio with MiniCPM-o 4.5 (#6628)

2026-08-31 10:27:34 +08:00

test_serve_cli.py

[Benchmark] Add local OmniInteract realtime benchmark (#6522)

2026-08-25 15:32:58 +00:00

test_stream_latency_metrics.py

[Edge][WP0] Add offline streaming-latency benchmark for chunked TTS

2026-09-09 10:24:53 +08:00