| { | |
| "audio_tokenizer": { | |
| "audio_tokenizer_class": "PocketTTSAudioTokenizerModel", | |
| "audio_tokenizer_name_or_path": "eustlb/pocket-tts-audio-tokenizer" | |
| }, | |
| "fade_in_duration": 0.005, | |
| "feature_extractor": { | |
| "analysis_frame_duration": 0.02, | |
| "end_on_pause": true, | |
| "fade_out_duration": 0.02, | |
| "feature_extractor_type": "PocketTTSFeatureExtractor", | |
| "feature_size": 1, | |
| "frame_size": 1920, | |
| "pad_to_multiple_of": 1920, | |
| "padding_side": "right", | |
| "padding_value": 0.0, | |
| "pause_duration": 0.08, | |
| "return_attention_mask": true, | |
| "sampling_rate": 24000, | |
| "silence_floor_db": 35.0 | |
| }, | |
| "frames_after_eos": 2, | |
| "frames_after_eos_short_text": 4, | |
| "processor_class": "PocketTTSProcessor", | |
| "short_text_max_words": 4 | |
| } | |