Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
@voicybot Telegram bot main repository
| Date | Stars |
|---|---|
| 2026-07-24 | 905 |
| 2026-07-25 | 905 |
| 2026-07-28 | 905 |
| 2026-07-30 | 905 |
| 2026-08-06 | 905 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
[](https://voicybot.com/) # [@voicybot](https://t.me/voicybot) main repository This repository contains the code for one of the most popular bots I've ever built for Telegram — [@voicybot](https://t.me/voicybot). It automatically converts all the audio messages to text when added to a group chat. Please, feel free to fork, add features and create pull requests so that everybody (over 12 000 000 people) can experience the features you've built. The active bot interface is maintained in English and Russian in the `locales` folder. ## List of repositories - [voicy](https://github.com/backmeupplz/voicy) — the main [@voicybot](https://t.me/voicybot) code - [voicy-payments](https://github.com/backmeupplz/voicy-payments) — payments service that used stripe to process payments for the Google Speech seconds of recognition; currently retired as the stats server for [voicybot.com](https://voicybot.com) - [voicy-landing](https://github.com/backmeupplz/voicy-landing) — [borodutch.com](https://borodutch.com) landing page - [voicy-recognition](https://github.com/backmeupplz/voicy-recognition/) — Recognition service for [voicybot.com](https://voicybot.com) ## Installation and local launch 1. Clone this repo: `git clone https://github.com/backmeupplz/voicy` 2. Launch a [mongo database](https://www.mongodb.com/) locally 3. Create `.env` file with the environment variables listed below 4. Install `ffmpeg` on machines that run local transcription workers 5. Run `yarn` in the root folder 6. Run `yarn start` ## Environment variables in `.env` file | Variable | Description | | ------------- | --------------------------------------------------------------- | | `MONGO` | URI for the mongo database used | | `TOKEN` | Telegram bot token | | `SALT` | Random salt to generate various encrypted stuff | | `ADMIN_ID` | Chat id of the person who shall receive valuable logs | | `ENVIRONMENT` | App environment, can be `development`, defaults to `production` | See examples in `.env.sample` file. Optional live activity stream: | Variable | Description | | -------------------------------------------- | ------------------------------------------------------------------------------------- | | `VOICY_ACTIVITY_STREAM_URL` | Symphony activity stream base URL or full `/activity/v1/events` URL | | `VOICY_ACTIVITY_STREAM_TOKEN` | Submit token sent as `Authorization: Bearer ...` | | `VOICY_ACTIVITY_STREAM_ANONYMIZATION_SECRET` | Optional HMAC secret for deterministic chat/user labels; defaults to the submit token | | `VOICY_ACTIVITY_STREAM_TIMEOUT_MS` | Optional submit timeout, default `750` | When the stream URL or token is missing, activity emitters are no-ops. Events are fire-and-forget and contain only anonymized chat labels, command names, media source kinds, and runtime/job milestones; Telegram message text, raw chat IDs, raw user IDs, file IDs, tokens, emails, and phone numbers must not be sent. ## Continuous integration Any commit pushed to `main` gets deployed to [@voicybot](https://t.me/voicybot) via [CI Ninja](https://github.com/backmeupplz/ci-ninja). ## Windows GPU transcription worker Voicy queues incoming Telegram audio in Mongo and expects one or more authenticated worker clients to process those jobs. The Windows GPU worker is the supported local setup for Nikita's RTX 4070 Ti machine: it polls the backend worker API, downloads the queued audio source, runs a local transcription command on CUDA, and uploads the final transcript back to Voicy. The worker API is mounted at `/worke
Excerpt of 11,444 characters
Read on GitHub
745
36
7
6
3
2
1
1
1
1
1
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:6450f3a927c6628c, topic:speech-to-text, readme:transcription