60 lines
1.4 KiB
JSON
60 lines
1.4 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": 17572.953226976097,
|
|
"playback_start_ms": 17572.95406376943,
|
|
"stall_at_ttfa_ms": 0,
|
|
"audio_s": 4.16,
|
|
"total_s": 18.025338406674564,
|
|
"rtf": 4.333014040066001,
|
|
"decode_ms_total": 1063.6401386000216,
|
|
"chunks": [
|
|
{
|
|
"idx": 0,
|
|
"t_ms": 17572.95406376943,
|
|
"arrival_ms": 16937.01417790726,
|
|
"frames": 25,
|
|
"samples": 48000,
|
|
"decode_ms": 635.8697768300772
|
|
},
|
|
{
|
|
"idx": 1,
|
|
"t_ms": 17849.87504594028,
|
|
"arrival_ms": 16937.01417790726,
|
|
"frames": 25,
|
|
"samples": 48000,
|
|
"decode_ms": 276.8944939598441
|
|
},
|
|
{
|
|
"idx": 2,
|
|
"t_ms": 18000.762850046158,
|
|
"arrival_ms": 16937.01417790726,
|
|
"frames": 2,
|
|
"samples": 3840,
|
|
"decode_ms": 150.87586781010032
|
|
}
|
|
],
|
|
"device": "cpu",
|
|
"wav": "/data/zhoutaichang/embedding_infer/vllm-omni/benchmarks/tts/edge_results/codec_stream/20260909-163326/client_0.wav",
|
|
"codes": "/data/zhoutaichang/embedding_infer/vllm-omni/benchmarks/tts/edge_results/codec_stream/20260909-163326/client_0.codes.npy"
|
|
} |