| { | |
| "audio_processor": { | |
| "eps": 1e-06, | |
| "feature_extractor_type": "VibeVoiceTokenizerProcessor", | |
| "normalize_audio": true, | |
| "sampling_rate": 24000, | |
| "target_dB_FS": -25 | |
| }, | |
| "db_normalize": true, | |
| "eps": 1e-06, | |
| "feature_extractor_type": "VibeVoiceAcousticTokenizerFeatureExtractor", | |
| "feature_size": 1, | |
| "language_model_pretrained_name": "Qwen/Qwen2.5-1.5B", | |
| "normalize_audio": true, | |
| "padding_side": "right", | |
| "padding_value": 0.0, | |
| "return_attention_mask": true, | |
| "sampling_rate": 24000, | |
| "speech_tok_compress_ratio": 3200, | |
| "target_dB_FS": -25 | |
| } | |