Skip to content
TomoriBot

KoboldCPP Transcription

Use your existing KoboldCPP instance to transcribe audio attachments and voice messages in TomoriBot.

KoboldCPP includes Whisper-based speech-to-text. TomoriBot connects to KoboldCPP using its OpenAI-compatible audio transcription endpoint (POST /v1/audio/transcriptions).

Setup

Section titled “Setup”

Start KoboldCPP with Whisper/STT enabled and confirm your build exposes:

  • POST /v1/audio/transcriptions
  • GET /v1/models or GET /models

Keep KoboldCPP running while TomoriBot is using it. If your build only exposes /api/extra/transcribe or another custom shape, use a wrapper until TomoriBot has a dedicated adapter.

Register in TomoriBot

Section titled “Register in TomoriBot”

Run /providers, choose Add New Custom Endpoint, and use the transcription API compatibility:

  • API Compatibility: openai-compatible-transcription
  • endpoint_url: your KoboldCPP server root

After saving the connection, select it and use its model dropdown to add the model name your server reports as a Transcription model.

Use /providers for endpoint registration and model setup. Then open /config > Models > Switch Models to select and activate the registered endpoint.

Use transcripts

Section titled “Use transcripts”

After registration, TomoriBot transcribes audio attachments in the background and adds the text to chat context. Use /config > Behavior > Notice Behavior only if you also want transcripts posted visibly in chat.