Skip to main content
The voice transcription plugin converts Discord voice messages - the audio clips members record directly in the Discord client - into text. When a member sends a voice message in a configured channel, the bot replies with a text transcription of what was said. This makes voice messages accessible to members who cannot or prefer not to listen to audio, improves searchability of conversations, and helps in noisy environments where listening is impractical.
This plugin transcribes Discord voice messages (the record-a-voice-message feature in the Discord client). It does not transcribe audio from voice channels.
Voice Transcription

​
Setup

Navigate to in the dashboard, or use the slash command:
Configure which channels the bot should watch for voice messages. You can enable transcription for specific channels or server-wide.
Shows which channels currently have transcription enabled.

​
How transcription works

When a voice message is sent in a configured channel, the bot downloads the audio, sends it to the Whisper speech-to-text model, and posts the transcript as a reply to the voice message. The transcript appears a few seconds after the voice message is sent, depending on the audio length. Transcriptions are clearly labelled as bot-generated and may contain errors, particularly for unusual accents, technical terms, or poor audio quality.

​
Dashboard configuration

Navigate to in the dashboard to:
  • Add or remove channels from the transcription list
  • View transcription history for a channel
  • Toggle transcription on or off without removing configuration

​
Required permissions