42 lines
1.7 KiB
JSON
42 lines
1.7 KiB
JSON
{
|
|
"model_details": {
|
|
"manufacturer_name": "Coqui",
|
|
"manufacturer_website": "https://github.com/coqui-ai/TTS and also updated engine https://github.com/idiap/coqui-ai-TTS/",
|
|
"model_description": "XTTS is a multi-speaker, voice cloning model. XTTS-v2 by Coqui AI is a voice generation model that lets you clone voices into a multitude of languages by using just a mere 6+ second audio sample. It is licenced under Coqui AI's Coqui Public Model License 1.0.0 which allows for `non-commercial use of a machine learning model and its outputs.`"
|
|
},
|
|
"model_capabilties": {
|
|
"audio_format": "wav",
|
|
"deepspeed_capable": true,
|
|
"generationspeed_capable": true,
|
|
"languages_capable": true,
|
|
"lowvram_capable": true,
|
|
"multimodel_capable": true,
|
|
"repetitionpenalty_capable": true,
|
|
"streaming_capable": true,
|
|
"temperature_capable": true,
|
|
"multivoice_capable": true,
|
|
"pitch_capable": false,
|
|
"windows_capable": true,
|
|
"linux_capable": true,
|
|
"mac_capable": true
|
|
},
|
|
"settings": {
|
|
"def_character_voice": "female_01.wav",
|
|
"def_narrator_voice": "male_01.wav",
|
|
"deepspeed_enabled": false,
|
|
"engine_installed": true,
|
|
"generationspeed_set": 1,
|
|
"lowvram_enabled": false,
|
|
"pitch_set": 0,
|
|
"repetitionpenalty_set": 10,
|
|
"temperature_set": 0.75
|
|
},
|
|
"openai_voices": {
|
|
"alloy": "female_01.wav",
|
|
"echo": "female_01.wav",
|
|
"fable": "female_01.wav",
|
|
"nova": "female_01.wav",
|
|
"onyx": "female_01.wav",
|
|
"shimmer": "female_01.wav"
|
|
}
|
|
} |