docs/source/templates/transcribe_audio.md
Listen to an audio file and transcribe its content in natural language, performing speech recognition.
!!! error Enterprise If you're managing more complex or high-volume audio labeling projects, Label Studio Enterprise includes an advanced audio transcription interface built to support faster, more precise annotation at scale.
See our new [Multi-Channel Audio Transcription](react_audio) template and learn more in [A New Audio Transcription UI for Speed and Quality at Scale](https://humansignal.com/blog/building-a-better-ui-for-audio-transcription-at-scale/) (blog post).
<View>
<Audio name="audio" value="$audio" zoom="true" hotkey="ctrl+enter" />
<Header value="Provide Transcription" />
<TextArea name="transcription" toName="audio" rows="4" editable="true" maxSubmissions="1" />
</View>
All labeling configurations must be wrapped in View tags.
Use the Audio object tag with the hotkey argument to allow annotators to play back audio on the labeling interface using a specific hotkey, and use the zoom="true" argument to allow annotators to zoom in on the audio wave:
<Audio name="audio" value="$audio" zoom="true" hotkey="ctrl+enter" />
You can add a header to provide instructions to the annotator:
<Header value="Provide Transcription" />
Use the TextArea control tag to prompt annotators to add a transcript for the audio:
<TextArea name="transcription" toName="audio" rows="4" editable="true" maxSubmissions="1" />
The rows="4" argument lets you configure the size of the text box visible on the labeling interface. The maxSubmissions="1" argument limits the maximum number of transcripts submitted by an annotator for the audio clip to one, while the editable="true" argument allows annotators to edit the transcript.