Repolode
제출하기
antirez

antirez/whisperbot

A Telegram bot written in C that turns voice messages into text using whisper.cpp or qwen-asr as the speech recognition backend.

GitHub에서 보기
스타
168
포크
12
열린 이슈
1
마지막 커밋
8개월 전

어떤 프로젝트인가요?

Whisperbot is a Telegram bot that transcribes voice messages. A user sends a voice message to the bot, which downloads the audio, converts it to WAV with ffmpeg, runs the chosen speech recognition backend as an external process, and streams the transcript back by editing the reply as text arrives.

The bot supports two backends: whisper.cpp (via whisper-cli) and qwen-asr, a pure C inference implementation of Qwen3-ASR. The backend is selected at compile time by defining one macro in whisperbot.c. It is built on botlib and uses a thread-per-request design with a mutex-guarded queue, since ASR is CPU-heavy. When the queue grows, the bot switches to a faster model, and it rejects new requests once the queue is full.

Limits such as maximum audio duration, queue size, and process timeout are set as constants at the top of the source. Backend paths and models are also hardcoded, and queued requests are lost on restart.

핵심 기능

  • 01Transcribes Telegram voice messages and replies with the text.
  • 02Supports whisper.cpp and qwen-asr as selectable ASR backends.
  • 03Streams partial transcription by editing the message as text arrives.
  • 04Switches to a faster model when the request queue grows.
  • 05Rejects new requests when the queue is full and asks users to retry later.

활용 사례

  • Running a personal or team voice-to-text bot on a small self-hosted server.
  • Transcribing voice notes sent in Telegram chats without sending audio to a third-party cloud service.
  • Experimenting with local speech recognition backends behind a chat interface.

토픽

#telegram-bot#speech-to-text#whisper#asr#voice-transcription

자주 묻는 질문

whisperbot은(는) 무엇인가요?
A Telegram bot written in C that turns voice messages into text using whisper.cpp or qwen-asr as the speech recognition backend. Whisperbot is a Telegram bot that transcribes voice messages. A user sends a voice message to the bot, which downloads the audio, converts it to WAV with ffmpeg, runs the chosen speech recognition backend as an external process, and streams the transcript back by editing the reply as text arrives.
whisperbot은(는) 무료로 사용할 수 있나요? 라이선스는 무엇인가요?
whisperbot에는 명시된 라이선스가 없습니다. 코드는 공개되어 있지만 라이선스가 없으면 재사용 허가가 명확하지 않으므로, 사용 전에 작성자에게 문의하는 것이 좋습니다.
whisperbot은(는) 아직 유지보수되고 있나요?
6개월 이상 커밋 없음. whisperbot의 마지막 커밋은 2026년 2월 12일입니다.
whisperbot은(는) 어디에 사용할 수 있나요?
Running a personal or team voice-to-text bot on a small self-hosted server. Transcribing voice notes sent in Telegram chats without sending audio to a third-party cloud service. Experimenting with local speech recognition backends behind a chat interface.
whisperbot은(는) 어디서 받을 수 있나요?
소스 코드와 설치 방법은 GitHub에 있습니다: https://github.com/antirez/whisperbot

요약은 프로젝트 README를 바탕으로 AI가 작성했습니다. GitHub 통계는 매일 갱신됩니다.

관련 프로젝트