How to Make Your First Audio Transcription - VoxScriber Tutorial

On the dashboard, click New transcription. On this screen you can:

• Drag and drop the file onto the indicated area;
• Select the file from your computer or phone;
• Record directly in the browser;
• Paste a YouTube link or media URL.

The upload shows the progress bar. Large files upload in chunks and, if the connection drops, the upload resumes from where it stopped — just upload the same file again.

Before starting, you can adjust:

• Transcription engine: the default is Premium (high accuracy, 4 cycles/min). Standard is more economical (1 cycle/min) and works well with noise; Nano runs in the browser and costs 0 cycles. See which engine to use.
• Language: choose the audio language or leave it on automatic detection.
• Speaker detection: turn it on to identify who is speaking. On the free plan, the screen shows up to 2 speakers.

Click start and wait. Typical times:

• Audio up to 5 minutes: ~25 to 30 seconds;
• Audio from 30 to 60 minutes: ~3 to 5 minutes;
• Audio over 1 hour: ~5 to 10 minutes.

You don't need to stay on the page — the transcription continues on the server and appears in your library when it finishes.

With the transcription ready, open it in your library. There you have:

• Player synchronized with the text;
• Editor to fix sections and rename speakers (saving is manual, via the save button);
• AI features: automatic title, summary, minutes, and translation;
• Export: on the free plan, PDF and TXT; paid plans unlock DOCX, SRT, VTT, and JSON.