Run GPT-SoVITS as API

Have you tried running the the GPT-SoVITS as an API? # WebAPI Documentation Run with: `python api_v2.py -a 127.0.0.1 -p 9880 -c GPT_SoVITS/configs/tts_infer.yaml` ## Parameters: - `-a` - Binding address, default "127.0.0.1" - `-p` - Binding port, default 9880 - `-c` - TTS config file path, default "GPT_SoVITS/configs/tts_infer.yaml" ## Endpoints: ### Text-to-Speech **Endpoint**: `/tts` GET Example: ``` http://127.0.0.1:9880/tts?text=example_text&text_lang=zh&ref_audio_path=archive_jingyuan_1.wav&prompt_lang=zh&prompt_text=example_prompt&text_split_method=cut5&batch_size=1&media_type=wav&streaming_mode=true ``` POST Body: ```json { "text": "", // Required: Text to synthesize "text_lang": "", // Required: Text language "ref_audio_path": "", // Required: Reference audio path "aux_ref_audio_paths": [], // Optional: Additional reference audios "prompt_text": "", // Optional: Prompt text "prompt_lang": "", // Required: Prompt language "top_k": 5, // Top k sampling "top_p": 1, // Top p sampling "temperature": 1, // Sampling temperature "text_split_method": "cut0", // Text segmentation method "batch_size": 1, // Inference batch size "batch_threshold": 0.75, // Batch splitting threshold "split_bucket": true, // Enable bucket splitting "speed_factor": 1.0, // Speech speed control "streaming_mode": false, // Enable streaming response "seed": -1, // Random seed "parallel_infer": true, // Enable parallel inference "repetition_penalty": 1.35 // T2S repetition penalty } ``` ### Control Commands **Endpoint**: `/control` Commands: - "restart": Restart service - "exit": Stop service GET Example: ``` http://127.0.0.1:9880/control?command=restart ``` ### Model Management **Set GPT Model**: ``` http://127.0.0.1:9880/set_gpt_weights?weights_path=GPT_SoVITS/pretrained_models/s1bert25hz-2kh-longer-epoch=68e-step=50232.ckpt ``` **Set SoVITS Model**: ``` http://127.0.0.1:9880/set_sovits_weights?weights_path=GPT_SoVITS/pretrained_models/s2G488k.pth ``` Responses: - Success: Returns "success" with HTTP 200 - Failure: Returns error message with HTTP 400 I translated that from this API file in the repository https://github.com/RVC-Boss/GPT-SoVITS/blob/main/api_v2.py