Definitions of key terms
- Project: The top-level container for a job, client, interview set, or research topic. See 4. Projects and how to create one.
- Media file: An audio or video file you upload for transcription. See 5. How to transcribe a file.
- Transcript: The editable text generated from a media file. See 6. Working with transcripts.
- Segment: One editable block of transcript text. In subtitle editor mode, each segment becomes a subtitle.
- Speaker label: A name such as Speaker 1, Interviewer, or Client. See 7. Guide to speaker recognition.
- Subproject: A project nested inside another project, useful for larger work. Creating subprojects requires a subscription.
- Note: A project note that opens in its own editor window.
- Queue: The list of media files waiting to be transcribed. See 5. How to transcribe a file.
- Model: The speech recognition model used for transcription: Whisper Base (free), or Whisper Small, Whisper Medium, Whisper Large Turbo and NVIDIA Parakeet v3 (subscription). See 8. Guide to Settings for model and dtype details.
- Dtype: The precision or quantization mode used when loading a model.
Looking for tags, advanced search or documents? These research tools are no longer part of Transcription App. They now live in Qualitativist, our qualitative research suite built on the same private, on-device transcription, with a literature library and citations in Word as well.