Telegram Voice To Text

Architecture Overview

Ready Speech to Text model
Ready puctuation model for text postprocessing
1. Ready russian model
2. Ready english model (if it works)
3. Finetune english model or train from scratch manually
Telegram bot as interface
Hosting as models' backend

Audio message summarizer.

We can enter text and get summarized post-processed output. Check raw text?

Name		Name	Last commit message	Last commit date
Latest commit History 13 Commits
handlers		handlers
keyboards		keyboards
ml		ml
utils/misc		utils/misc
.gitignore		.gitignore
audio.zip		audio.zip
loader.py		loader.py
main.py		main.py
readme.md		readme.md
requirements.txt		requirements.txt
runtime.txt		runtime.txt
transcriptions.csv		transcriptions.csv