transcribe

To transcribe audio with speaker labels, you can convert spoken content into searchable text while identifying different speakers. This approach uses OpenAI's Whisper models to generate text and applies speaker diarization to attribute dialogue to specific individuals.

Speaker diarization is the process of identifying and segmenting different speakers in an audio recording. For meeting transcription, it labels individual attendees' contributions, allowing you to attribute comments accurately and generate structured meeting minutes.

Yes, you need an OpenAI API key and a Python environment with the openai SDK installed to convert speech to text. This Skill relies on OpenAI's Whisper models for accurate transcription and requires this setup to process your audio or video files.

Yes, you can transcribe video recordings for content analysis. The Skill processes the audio track from video files to convert speech into text, making the spoken content searchable and editable for further analysis, such as extracting action items.

Available output formats for audio transcription include plain text, JSON, and diarized JSON. These formats allow you to choose between simple readability and structured data that retains speaker identification for downstream processing.

To identify known speakers in an interview transcription, the Skill supports known speaker identification alongside speaker diarization. This allows you to match diarized audio segments to specific participant names, creating an accurate interview record.