There’s a lot of decent models out there, these days. The following are all solid options:
- F5-TTS (https://github.com/SWivid/F5-TTS.git)
- Sesame CSM ([email protected]:SesameAILabs/csm.git)
- SparkTTS (https://github.com/SparkAudio/Spark-TTS)
- Kokoro (https://github.com/hexgrad/kokoro)
0.9438Hz, i.e. 1*(2^(-100/1200)) IIRC.