search

rksm / org-ai

Emacs as your personal AI assistant. Use LLMs such as ChatGPT or LLaMA for text generation or DALL-E and Stable Diffusion for image generation. Also supports speech input / output.
GNU General Public License v3.0
655 stars 53 forks source link

espeak, instead of whisper #49

Open OrionRandD opened 1 year ago

OrionRandD commented 1 year ago

Not an issue, but a suggestion: I tried to use whisper. But, since I am using Linux, I prefer using "espeak" and all wonderful resources it provides... Many voices, phonology output, et.al.

I just select the text I want to hear and issue: :!espeak -ven-gb
and hear the robotized voice of espeak... If I want to record the text I can redirect the stdout to an ogg/mp3/wav file...

rksm commented 1 year ago

speech synthesis using espeak (by means of greader) is already supported. I've now documented it here: https://github.com/rksm/org-ai#espeak--greader Basically, just install the greader package from elpa and it should work.

I'm not aware that espeak is able to do speech-to-text, though. Am I missing something?