|
ailia_voice
1.5.0.0
|


Public Member Functions | |
| int | GetEnvironmentId (bool is_gpu) |
| Get the environmen id. More... | |
| string | GetEnvironmentName () |
| Get the environmen name. More... | |
| bool | Create (int env_id, int flag) |
| Create a instance. More... | |
| bool | SetUserDictionary (string dict_path, int dict_type) |
| Set user dictionary file. More... | |
| bool | OpenDictionary (string dict_path, int dict_type) |
| Set dictionary path. More... | |
| bool | OpenModel (string encoder, string decoder1, string decoder2, string wave, string ssl, int model_type, int cleaner_type) |
| bool | OpenTacotron2ModelFile (string encoder, string decoder1, string decoder2, string wave, int cleaner_type) |
| Open Tacotron2 model. More... | |
| bool | OpenGPTSoVITSV1ModelFile (string encoder, string decoder1, string decoder2, string wave, string ssl) |
| Open GPT-SoVITS V1 model. More... | |
| bool | OpenGPTSoVITSV2ModelFile (string encoder, string decoder1, string decoder2, string wave, string ssl, string chinese_bert, string vocab) |
| Open GPT-SoVITS V2 model. More... | |
| bool | OpenGPTSoVITSV3ModelFile (string encoder, string decoder1, string decoder2, string ssl, string vq, string cfm, string bigvgan, string chinese_bert, string vocab) |
| Open GPT-SoVITS V3 model. More... | |
| bool | OpenGPTSoVITSV2ProModelFile (string encoder, string decoder1, string decoder2, string ssl, string vits, string sv, string chinese_bert, string vocab) |
| Open GPT-SoVITS V2-Pro model. More... | |
| bool | SetSampleSteps (int steps) |
| Set CFM sampling steps. More... | |
| bool | SetSpeed (float speed) |
| Set the speech speed for synthesis. More... | |
| bool | SetModelType (int model_type) |
| Set the model type for G2P processing. More... | |
| virtual void | Close () |
| Destroys instance. More... | |
| virtual void | Dispose () |
| Release resources. More... | |
| string | G2P (string utf8, int g2p_type) |
| Get features. More... | |
| bool | SetReference (AudioClip ref_audio, string ref_text) |
| Set reference audio. More... | |
| bool | Inference (string feature) |
| Perform inference. More... | |
| AudioClip | GetAudioClip () |
| Get audio clip. More... | |
Protected Member Functions | |
| virtual void | Dispose (bool disposing) |
|
inlinevirtual |
|
inline |
Create a instance.
| env_id | Environment ID of ailia |
| flag | OR of flags (AiliaVoice.AILIA_VOICE_FLAG_*) |
|
inlinevirtual |
Release resources.
|
inlineprotectedvirtual |
|
inline |
Get features.
| utf8 | Input string |
| g2p_type | G2P type |
|
inline |
|
inline |
Get the environmen id.
| is_gpu | Whether to use GPU |
|
inline |
|
inline |
Perform inference.
| feature | Input feature string |
|
inline |
Set dictionary path.
| net | A network instance pointer |
| dictionary_path | The path name to the dictionary folder |
| dictionary_type | AILIA_VOICE_DICTIONARY_TYPE_* |
|
inline |
Open GPT-SoVITS V1 model.
| encoder | The path name to the onnx file |
| decoder1 | The path name to the onnx file |
| decoder2 | The path name to the onnx file |
| wave | The path name to the onnx file |
| ssl | The path name to the onnx file |
|
inline |
Open GPT-SoVITS V2 model.
| encoder | The path name to the onnx file |
| decoder1 | The path name to the onnx file |
| decoder2 | The path name to the onnx file |
| wave | The path name to the onnx file |
| ssl | The path name to the onnx file |
| chinese_bert | The path name to the chinese-roberta.onnx file (null to disable Chinese BERT) |
| vocab | The path name to the vocab.txt file (null to disable Chinese BERT) |
|
inline |
Open GPT-SoVITS V2-Pro model.
| encoder | The path name to the onnx file |
| decoder1 | The path name to the onnx file |
| decoder2 | The path name to the onnx file |
| ssl | The path name to the onnx file |
| vits | The path name to the onnx file |
| sv | The path name to the onnx file |
| chinese_bert | The path name to the chinese-roberta.onnx file (null to disable Chinese BERT) |
| vocab | The path name to the vocab.txt file (null to disable Chinese BERT) |
|
inline |
Open GPT-SoVITS V3 model.
| encoder | The path name to the onnx file |
| decoder1 | The path name to the onnx file |
| decoder2 | The path name to the onnx file |
| ssl | The path name to the onnx file |
| vq | The path name to the onnx file |
| cfm | The path name to the onnx file |
| bigvgan | The path name to the onnx file |
| chinese_bert | The path name to the chinese-roberta.onnx file (null to disable Chinese BERT) |
| vocab | The path name to the vocab.txt file (null to disable Chinese BERT) |
|
inline |
| net | A network instance pointer |
| encoder | The path name to the onnx file |
| decoder1 | The path name to the onnx file |
| decoder2 | The path name to the onnx file |
| wave | The path name to the onnx file |
| ssl | The path name to the onnx file |
| model_type | AILIA_VOICE_MODEL_TYPE_* |
| cleaner_type | AILIA_VOICE_CLEANER_TYPE_* |
|
inline |
Open Tacotron2 model.
| encoder | The path name to the onnx file |
| decoder1 | The path name to the onnx file |
| decoder2 | The path name to the onnx file |
| wave | The path name to the onnx file |
| cleaner_type | AILIA_VOICE_CLEANER_TYPE_* |
|
inline |
Set the model type for G2P processing.
| model_type | AILIA_VOICE_MODEL_TYPE_* |
Sets the model type when using G2P standalone without loading model files. Automatically set when OpenModel or OpenGPTSoVITSV3ModelFile is called.
|
inline |
Set reference audio.
| ref_audio | Reference audio |
| ref_text | Reference text |
|
inline |
Set CFM sampling steps.
| steps | Number of sampling steps |
|
inline |
Set the speech speed for synthesis.
| speed | Speed value (default 1.0, must be greater than 0) |
Supported by GPT-SoVITS V2 and V3. Not effective for V1.
|
inline |
Set user dictionary file.
| net | A network instance pointer |
| dictionary_path | The path name to the user dictionary file |
| dictionary_type | AILIA_VOICE_DICTIONARY_TYPE_* |
You need to call before OpenDictionary.