Larynx

Generates Audiobooks from PDF files

LLaMA

experiment with LLaMA. Create a voice chat engine "maybe Offine" with some TTS and some LLM model. LLaMA has some kind of waiting list I Guess ! maybe some other model ?? do it with semantic search engine. It may help !

generate music as well

Research or Documenting :

1. Train wavenet and try to convert some text files to audio (upload the results to youtube and twitter)
2. next step would be to train some better models that can convert to audio on small samples
3. LLaMA can help with some chat features not more then that make a seperate repo for that kinda stuff
4. Generate music with advances TTS models with small sample size requirements

with 1 research on how to convert pdf files to normal text files !
for wavenet - how to convert specgrams to audio files ?