transcribe

Transcribe audio with speaker diarization by running the bundled CLI on meeting or interview recordings to generate text labeled by who spoke when. You can optionally provide up to four known-speaker references to accurately identify participants.

Speaker diarization identifies who spoke when in audio transcripts. You need it for meetings, interviews, and media recordings where distinguishing individual speakers is essential for accurate documentation and contextual understanding.

Yes, you need an OPENAI_API_KEY to transcribe audio files because the Skill makes live calls to OpenAI for speech-to-text processing. This key enables the transcription and optional diarization features.

Yes, you can use known-speaker hints to identify speakers in interview transcripts. The Skill supports up to four known-speaker references to accurately label participants during the diarization process.

The best way to handle long audio recordings for speech-to-text conversion is using the Skill's chunking strategy. This approach splits extended media into manageable segments for repeatable, deterministic transcription runs.

The Skill supports multiple output formats when transcribing audio to text, allowing you to format meeting and interview transcripts for documentation, indexing, or further processing as needed.