Tutorial 16cGoing further

Transcribing audio

Turn a recording into text with xAI STT v1, with speaker labels and filler words when you need them.

00:00 / 01:06

What this video covers

In this part, we're transcribing a recording. Pick xAI STT v1 in the model picker.

This model doesn't take a prompt: the prompt box is disabled, and an Audio File slot opens under Reference.

You can upload a file from your computer, or right-click one in your library and choose Use as Reference. That's how we add the voice-over from the first part.

Once the file is in, the price appears; it's based on how long the file is.

The language is detected automatically, or you can pick it from the list yourself.

If more than one person is speaking, switch on Speaker labels. To keep fillers like "um" and "uh" in the text, switch on Keep filler words.

Press Generate. The transcript arrives in seconds, shown with its word count.

Press Copy, and the text goes to your clipboard.

The transcript is saved to your library, too.

To sum up: for voice-over, write the text; for music, describe the style; for a transcript, give it an audio file.

Try it yourself

Create a free LUVI account and follow along with the video.

Get started

More from Going further

All tutorials