Transcribe Audio to Text

Turn any recording into text, with automatic language detection across 90+ languages.

Audio
Cost—

105 credits per minute · 1 transcript · A few seconds per minute of audio

Related

795
Product Ad

Turn a product photo into a short, creator-style ad for social.

288
Stem Splitter

Split any song into six stems for remixing, sampling or practice.

129
Thumbnail Maker

Scroll-stopping video thumbnails from a title and a face.

Extract Audio from Video

Pull the sound track out of any video as MP3 or WAV. Free.

Vocal Remover

Split a song into a clean vocal track and an instrumental, for karaoke, remixes and practice.

Background Noise Remover

Clean up speech recordings: remove hiss, hum, traffic and room noise, then even out the loudness.

Text to Speech

Natural, expressive speech from a script. Pick a voice and direct the delivery with tags like [whispering].

Create Music

Original songs with vocals, or instrumentals, from a description of the style and mood.

Video to GIF

Turn the start of any video into a looping GIF. Pick the length, size and frame rate. Free.

Add Subtitles to Video

Transcribe a video and burn styled subtitles into it automatically. Pick a style and position.

Video Avatar

Make a photo talk: add a portrait and a voice clip, get a lip-synced video of them speaking.

3D Cartoon

Animated-feature style 3D scenes from a line of text: expressive characters, soft light, bright color.

Recast Video

Swap who's on screen: your character performs the exact moves and expressions of a reference video.

Lipsync Video

Make the person in a video say something new: their lips follow the voice clip you add.

Remove Video Background

Cut the person or subject out of a video. Transparent WebM, or a green, white or black backdrop.

90s Anime

Hand-drawn 1990s anime: cel shading, soft film grain and the washed-out colors of the VHS era.

Vintage Cartoon

1930s rubber-hose animation: bouncy black-and-white characters, film flicker and grain.

Interview Film

A documentary-style interview shot from one photo: the person speaks your lines, seated, softly lit.

Product Loop Video

Turn a product photo into a seamless looping video: the product turns slowly in a clean studio.

American Comic

Comic-book panels with bold inks, halftone dots, dynamic angles and speech bubbles.

Nine Expressions

One photo in, a 3×3 grid of nine expressions out: happy, sad, surprised, laughing and more.

Reverse Video

Play any clip backwards, with the sound reversed too or muted. Free, MP4 out.

GIF Maker

Turn 2 to 10 pictures into an animated GIF. Choose the speed and size. Free, no watermark.

Compress GIF

Make animated GIFs smaller for chat, email and the web. Every frame kept. Free.

Explore by category

Popular

Newest

Transcribe audio to text

Type
File app
Input
Audio
Output
Transcript
Options
10 languages
Price
105 credits per minute
Plan
All plans, including Free

Upload a recording and read it a few seconds later. Interviews, meetings, lectures, voice notes and podcasts become punctuated text you can copy, search and quote.

How to transcribe audio

  1. Add the recording: MP3 or WAV, up to 60 minutes.
  2. Pick the language, or leave it on Auto-detect.
  3. Generate. A few seconds per minute of audio.
  4. Copy the text or download the .txt file.

Uses

  • Interviews and research you need to quote accurately.
  • Meeting notes without typing.
  • Show notes and articles from podcasts and videos.
  • Subtitles: a transcript is the first step to captions.

Tips

  • Clean noisy audio first with the Background Noise Remover.
  • Set the language for short clips; auto-detection has less to go on.
  • Working from a video? Extract its audio first, then transcribe the MP3.

FAQ

Which languages are supported?

More than 90. Leave Language on Auto-detect, or pick English, Spanish, French, German, Portuguese, Italian, Japanese, Korean or Chinese when you know it.

What do I get?

A plain-text transcript with punctuation, shown in full with a copy button and saved as a .txt file in your library.

Does it include timestamps or speaker names?

Not yet. You get the text in order, without timestamps or speaker labels.

How much does it cost?

105 credits per minute of audio, billed by the second, with a 10-credit minimum. Recordings can be up to 60 minutes.

How accurate is it?

Very accurate on clear speech. Background noise, overlapping voices and strong accents lower accuracy; clean the audio first if it's noisy.