| { | |
| "engine": "openvoice-v2", | |
| "components": { | |
| "tone_ref_encoder": "tone_ref_encoder.onnx", | |
| "tone_ref_encoder_q8": "tone_ref_encoder_q8.onnx", | |
| "tone_converter": "tone_converter.onnx", | |
| "tone_converter_q8": "tone_converter_q8.onnx" | |
| }, | |
| "sample_rates": { | |
| "input": 22050, | |
| "output": 22050 | |
| }, | |
| "distributable": true, | |
| "metadata": { | |
| "opset": 14, | |
| "spec_channels": 513, | |
| "tone_embedding_dim": 256, | |
| "upstream_hf": "myshell-ai/OpenVoiceV2", | |
| "license": "MIT", | |
| "export_path": "upstream SynthesizerTrn (no reconstruction)", | |
| "ref_enc_input": "linear_spectrogram (B, T, 513)", | |
| "converter_input": "linear_spectrogram (B, 513, T)", | |
| "converter_output": "raw_waveform (B, 1, samples)" | |
| } | |
| } |