[Bug] ValueError running YourTTS model on tts-server

🐛 Description

I am trying to run the YourTTS model on the tts-server. I am hitting Internal server error. When I check the log I see ValueError: [!] Look like you use a multi-lingual model. You need to define either a language_name or a style_wav to use a multi-lingual model.

To Reproduce

Here is the command: tts-server --model_name tts_models/multilingual/multi-dataset/your_tts

Environment

• 🐸TTS Version (e.g., 1.3.0): 0.5.0
• PyTorch Version (e.g., 1.8): 1.10.0
• Python version: 3.7.10
• OS (e.g., Linux): Linux
• CUDA/cuDNN version: None
• GPU models and configuration: None
• How you installed PyTorch (conda, pip, source): pip
• Any other relevant information:

Additional context

Here is the log

 " [!] Look like you use a multi-lingual model. "                             
ValueError:  [!] Look like you use a multi-lingual model. You need to define either a language_name or a style_wav to use a multi-lingual model.                                                                          
::1 - - [06/Jan/2022 23:14:17] "GET /api/tts?text=hi&speaker_id=male-pt-3%0A&style_wav= HTTP/1.1" 500 -  

Other models work fine. The problem looks like it is only for YourTTS model

[godspirit00]

I think it is because the server.py @ L176 does not pass a language_name to synthesizer.tts() just like synthesize.py @ L287.

So to fix this, just pass the language_name like en to synthesizer.tts.

    wavs = synthesizer.tts(text, speaker_name=speaker_idx, language_name="en", style_wav=style_wav)

[erogol]

server.py does not support YourTTS fully yet.