A LocalLLaMA community thread invites users to share their preferred open-weight or open-source audio models for speech recognition, text-to-speech and text-to-music, with details about testing and real-world use.
Key points
The thread asks for recommendations across ASR, TTS, STT and text-to-music models
Only open-weight or open-source models are eligible
Users are asked to describe their setup, tools and personal or professional use