Scribe
Record meetings and conversations, then get full transcripts with speaker identification.
Scribe is Ottex's meeting recorder. Hit record, talk, and get a searchable transcript organized by speaker — all processed locally or through your configured AI providers.
Recording a Meeting
- 1Open Ottex and go to the Scribe section in the sidebar
- 2Click "Record Meeting" in the top right
- 3Ottex starts capturing audio from your microphone (and system audio if configured)
- 4Speak naturally — Ottex transcribes in real-time
- 5Click Stop when you're done
- 6Your transcript appears organized by time and speaker
Understanding Transcripts
Each recording shows key metadata at the top:
The transcript text shows what was said in chronological order. When speaker diarization is enabled, each speaker's words are labeled separately.
Organizing with Projects
Projects help you keep recordings organized. You can create projects for different clients, topics, or workflows.
- 1In the Scribe sidebar, click "+ New Project"
- 2Give your project a name
- 3New recordings can be assigned to a project
- 4Use the sidebar to filter recordings by project
Importing Audio Files
You don't have to record live — Scribe can also transcribe existing audio files. Drag and drop an audio file into the drop zone at the bottom of the Scribe view, or click it to browse for a file.
Supported formats include MP3, WAV, M4A, and other common audio formats.
Speaker Diarization
When recording meetings with multiple participants, Ottex can identify different speakers and label their segments. This requires a provider that supports diarization (like Soniox or Deepgram).
Speaker labels appear as "Speaker 1", "Speaker 2", etc. in the transcript. The number of detected speakers is shown in the recording metadata.
Scribe Settings
Click the gear icon next to "Scribe" in the header to access settings specific to meeting recording:
- Recording model — which AI model to use for meeting transcription
- Audio source — microphone only, or also capture system audio
- Auto-save — automatically save recordings when you stop
- Speaker detection — enable/disable speaker diarization