Audio Transcription

A recording is only useful when people can find and review its transcript. I built the React frontend from scratch, bringing recording, transcript editing, and summaries into one workflow through REST APIs.

A recording passes through capture, processing, and review before it becomes usable documentation. Users need to see its current state and find the associated transcript and summary without losing context.

I built the React frontend from scratch so users could record audio, follow its processing status, and review the transcript and summary in one place. I connected those screens to REST APIs and handled the recording and review states. The audio-processing and summary services were outside my frontend work.

• Browser recording

  Capture and manage audio directly in the application.

• Transcript review

  View, edit, and manage transcripts after processing.

• Documentation summaries

  Manage summaries alongside their recordings and transcripts.

• Role-based access

  Control which recordings, transcripts, and summaries a user can access.

I integrated browser recording with transcript and summary management so users could complete the workflow in one application. This required handling microphone permissions, recording states, and failures alongside the review interface.

Clinical teams can move from recording to reviewing documentation in one application, with access controlled by role.

• Microphone Permissions Belong in the Recording State Machine →
• Six States Every Browser Audio Recorder Should Handle →
• From Audio to Insight: ASR and Diarization Pipelines →