transcribe

To transcribe audio with speaker diarization, you can use a bundled CLI that processes recordings and outputs labeled text. It identifies who spoke when, making it ideal for interviews, meetings, and podcasts requiring accurate speaker attribution.

Audio transcription supports multiple output formats including plain text, JSON, and diarized JSON. The diarized JSON format includes speaker labels and timestamps, providing structured data for downstream processing of interviews or meetings.

Yes, an OpenAI API key is required. The transcription process validates your OPENAI_API_KEY before converting audio to text and performing speaker diarization, ensuring secure access to the underlying speech-to-text processing.

Yes, you can provide known-speaker references to the transcription process. This allows the system to accurately label transcripts with specific speaker names, improving readability for meetings, podcasts, and interviews where participants are known.

You can optimize speech-to-text processing by providing optional language hints and applying chunking strategies. These features help the transcription engine handle diverse audio recordings more accurately by segmenting audio and guiding language recognition.

The best way to transcribe podcasts is using a CLI tool that combines speech-to-text conversion with optional speaker diarization. This approach handles long recordings through chunking and outputs structured formats like diarized JSON for easy editing.