1# Clone the server repo2git clone https://github.com/GoodGangLabs/cosyvoice3-instruct-multilanguage-multispeaker-server.git
34# Download model from HuggingFace5huggingface-cli download gglabs/cosyvoice3-multilingual-multispeaker-v3_ep49 --local-dir ./models/v3_ep49
67# Run with Docker8docker-compose up -d
910# Test TTS11curl -X POST http://localhost:8090/tts \12 -H "Content-Type: application/json"\13 -d '{"text": "안녕하세요", "language": "ko", "spk_id": "nalnani"}'
Direct Python Usage
python
1from cosyvoice.cli.cosyvoice import CosyVoice3
23# Load model4model = CosyVoice3("./models/v3_ep49", load_vllm=True)56# Generate speech7for result in model.inference_instruct2(8"안녕하세요, 반갑습니다.",9"한국어로 자연스럽게 말해주세요.",10"nalnani"11):12# result contains audio waveform13pass