Descript supports automatic transcription in 23 languages. You can import, transcribe, and quickly apply speaker labels to files with one speaker, or use speaker detection for files with multiple speakers. In this article and video, we'll show you how automatic transcription works for imported files.
The first time you import and transcribe files in Descript, you will be prompted to set the transcription language. We recommend enabling “Always ask before transcribing” if you plan on transcribing in different languages from project to project. You can always adjust it later from your preferences.
Importing and transcribing a single file
- Open or create a new project.
- Drag and drop your video or audio file into the Script Editor.
- If you add the file to a new blank composition, drag and drop the file anywhere in the Script Editor.
- If you are adding the file to a composition that already contains transcribed media:
- Place your cursor at the end of a paragraph where you want to add the file and click Enter to create a new blank line. If you're going to add media to the middle of a paragraph or sentence, click Enter twice.
- Drag and drop the file onto the new blank line in your Script Editor.
- Enter the speaker’s name. If you have more than one speaker in the file, set the number of speakers to detect to the right of the speaker's name.
- Click Done. If you selected more than one speaker, select Identify speakers in the message in the bottom right corner of the app. Listen to portions of your files, type in the speaker's name, or select the speaker’s name from the drop-down menu.
Importing and transcribing multiple files
- Open or create a new project.
- Drag and drop media files into a new composition together simultaneously.
You can create a sequence or add each transcribed file one after the other.
- Enter the speaker’s name. If you have more than one speaker in the file, set the number of speakers to detect to the right of the speaker's name.
- Click Done. If you choose to detect speakers, select Identify speakers in the message in the bottom right corner of the app. Listen to portions of your files, type in the speaker's name, or select the speaker’s name from the drop-down menu.
- If you need to add more interviews or recordings, you can create a new composition and repeat the process above.