59 lines
1.3 KiB
JSON
59 lines
1.3 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": 18017.76840770617,
|
|
"playback_start_ms": 18017.76904705912,
|
|
"stall_at_ttfa_ms": 0,
|
|
"audio_s": 4.16,
|
|
"total_s": 18.459118940867484,
|
|
"rtf": 4.437288206939299,
|
|
"decode_ms_total": 1080.4549525491893,
|
|
"chunks": [
|
|
{
|
|
"idx": 0,
|
|
"t_ms": 18017.76904705912,
|
|
"arrival_ms": 17375.3689751029,
|
|
"frames": 25,
|
|
"samples": 48000,
|
|
"decode_ms": 642.2775778919458
|
|
},
|
|
{
|
|
"idx": 1,
|
|
"t_ms": 18296.088204719126,
|
|
"arrival_ms": 17375.3689751029,
|
|
"frames": 25,
|
|
"samples": 48000,
|
|
"decode_ms": 278.30050280317664
|
|
},
|
|
{
|
|
"idx": 2,
|
|
"t_ms": 18455.97896585241,
|
|
"arrival_ms": 17375.3689751029,
|
|
"frames": 2,
|
|
"samples": 3840,
|
|
"decode_ms": 159.87687185406685
|
|
}
|
|
],
|
|
"device": "cpu",
|
|
"wav": "/data/zhoutaichang/embedding_infer/vllm-omni/benchmarks/tts/edge_results/codec_stream/20260909-154659/client_0.wav"
|
|
} |