CosyVoice open-source address: https://github.com/FunAudioLLM/CosyVoice

CosyVoice3's official webui.py cannot be used for integration because its audio component is streaming, causing API calls to return m3u8 files instead of wav audio. You need to open the webui.py file, search for the code streaming=True, change it to streaming=False, and then restart webui.py

---

Using in Video Translation Software

Follow these steps:

1. Deploy the official project, ensure webui.py launches and you've successfully completed a voice dubbing operation in the UI. Then download the modified webui.py file, overwrite the official one, and restart. Download link: https://github.com/jianchang512/stt/releases/download/0.0/cosyvoice3-webui-py.zip

2. For Windows systems, use the integrated package directly:

   • Download link (Baidu Netdisk): https://pan.baidu.com/s/1Lp1GNGZGaO6WQG3NLli_RQ?pwd=1234
   • Download link (Hugging Face): https://huggingface.co/mortimerme/repocollect/resolve/main/cosyvoice3-0529.7z?download=true

Launch and Use in pyVideoTrans

1. First, upgrade the pyVideoTrans software to the latest version.

2. Ensure the CosyVoice project is deployed and webui.py is running. You should be able to open http://127.0.0.1:8000 in your browser.

3. Open the video translation software, go to Settings in the top left corner -- CosyVoice: Enter the webui.py address, which defaults to http://127.0.0.1:8000.

4. Enter the reference audio and its corresponding text.

5. After filling in the details, on the main interface, select CosyVoice as the dubbing channel and choose the corresponding character. The clone character replicates the voice from the original video.

Important Notes

• The first use will automatically download the model from modelscope.cn, which may take some time. Please be patient.