76 lines
1.8 KiB
JSON
76 lines
1.8 KiB
JSON
{
|
|
"codec_start": {
|
|
"type": "codec.start",
|
|
"version": 1,
|
|
"utterance_index": 0,
|
|
"sentence_index": 0,
|
|
"sentence_text": "Hello, welcome to the voice synthesis benchmark test.",
|
|
"model": "Qwen/Qwen3-TTS-12Hz-0.6B-CustomVoice",
|
|
"codebooks": 16,
|
|
"codebook_size": 2048,
|
|
"frame_rate_hz": 12.5,
|
|
"sample_rate": 24000,
|
|
"decoder_id": "qwen3_tts_tokenizer_v2_12hz",
|
|
"header_format": "<IHB"
|
|
},
|
|
"schedule": [
|
|
2,
|
|
4,
|
|
8,
|
|
16,
|
|
25
|
|
],
|
|
"total_frames": 52,
|
|
"ttfa_ms": 15033.681467641145,
|
|
"playback_start_ms": 15554.042749684304,
|
|
"stall_at_ttfa_ms": 520.3603609651327,
|
|
"audio_s": 4.16,
|
|
"total_s": 17.163251301739365,
|
|
"rtf": 4.1257815629181165,
|
|
"decode_ms_total": 1315.1106755249202,
|
|
"chunks": [
|
|
{
|
|
"idx": 0,
|
|
"t_ms": 15033.682388719171,
|
|
"arrival_ms": 14931.589820887893,
|
|
"frames": 2,
|
|
"samples": 3840,
|
|
"decode_ms": 102.06782910972834
|
|
},
|
|
{
|
|
"idx": 1,
|
|
"t_ms": 15224.257772788405,
|
|
"arrival_ms": 15098.3333629556,
|
|
"frames": 4,
|
|
"samples": 7680,
|
|
"decode_ms": 125.90203760191798
|
|
},
|
|
{
|
|
"idx": 2,
|
|
"t_ms": 15575.003643985838,
|
|
"arrival_ms": 15412.30756463483,
|
|
"frames": 8,
|
|
"samples": 15360,
|
|
"decode_ms": 162.672720849514
|
|
},
|
|
{
|
|
"idx": 3,
|
|
"t_ms": 16674.042749684304,
|
|
"arrival_ms": 16016.14229567349,
|
|
"frames": 16,
|
|
"samples": 30720,
|
|
"decode_ms": 657.8587740659714
|
|
},
|
|
{
|
|
"idx": 4,
|
|
"t_ms": 17139.154572039843,
|
|
"arrival_ms": 16872.522183693945,
|
|
"frames": 22,
|
|
"samples": 42240,
|
|
"decode_ms": 266.6093138977885
|
|
}
|
|
],
|
|
"device": "cpu",
|
|
"wav": "/data/zhoutaichang/embedding_infer/vllm-omni/benchmarks/tts/edge_results/codec_stream/20260909-164759/client_0.wav",
|
|
"codes": "/data/zhoutaichang/embedding_infer/vllm-omni/benchmarks/tts/edge_results/codec_stream/20260909-164759/client_0.codes.npy"
|
|
} |