I tried the english TTS example and the result is quite underwhelming (compared to bark or polly/azure-tts ). It sounds like TTS systems one or two decades ago.

I tried the english TTS example and the result is quite underwhelming (compared to bark or polly/azure-tts ). It sounds like TTS systems one or two decades ago. Would those language-specific TTS models need to be finetuned?