Re:Abogen – Generate audiobooks from EPUBs, PDFs and text
by leobg on Sunday August 10, 2025 @06:23AM
(#44854182)
Kokoro is small and fast because all the text -> phoneme conversion is done by “dumb code” and only the phoneme -> sound part is done using a neural net.
The difference is that even weak LLMs are good at magically doing this, so I wonder what the problem is for the TTS mentioned above.