{
  "model_path": "models/Qwen3-TTS-12Hz-0.6B-Base",
  "ref_audio_path": "examples/readme_clone_input.wav",
  "ref_text": "Okay. Yeah. I resent you. I love you. I respect you. But you know what? You blew it! And thanks to you.",
  "synth_text": "Hello.",
  "outputs": [
    {
      "name": "text_tokens",
      "path": "text_tokens.bin",
      "shape": [
        1,
        10
      ],
      "dtype": "int64",
      "size_bytes": 80
    },
    {
      "name": "ref_text_tokens",
      "path": "ref_text_tokens.bin",
      "shape": [
        1,
        35
      ],
      "dtype": "int64",
      "size_bytes": 280
    },
    {
      "name": "ref_audio_embedding",
      "path": "ref_audio_embedding.bin",
      "shape": [
        1024
      ],
      "dtype": "float32",
      "size_bytes": 4096
    },
    {
      "name": "speech_codes",
      "path": "speech_codes.bin",
      "shape": [
        63,
        16
      ],
      "dtype": "int64",
      "size_bytes": 8064
    },
    {
      "name": "hidden_states",
      "path": "hidden_states.bin",
      "shape": [
        63,
        1024
      ],
      "dtype": "float32",
      "size_bytes": 258048
    },
    {
      "name": "decoded_audio",
      "path": "decoded_audio.bin",
      "shape": [
        120960
      ],
      "dtype": "float32",
      "size_bytes": 483840,
      "sample_rate": 24000,
      "duration_seconds": 5.04
    },
    {
      "name": "output_wav",
      "path": "output.wav",
      "sample_rate": 24000,
      "duration_seconds": 5.04,
      "format": "WAV PCM float32"
    }
  ],
  "model_info": {
    "tts_model_type": "base",
    "tokenizer_type": "qwen3_tts_tokenizer_12hz",
    "device": "cpu",
    "dtype": "bfloat16"
  }
}