Convert audio to text, even a 3-hour recording
Upload a meeting, interview or lecture, or paste a link. You get the full transcript, who said what, and a summary.
No account? The first 30 minutes of any file are free. Create a free account for premium on your first file.
- 4.8from 1,690 reviews
- 6,730+ accounts created
- 717,311+ minutes transcribed
Thanks for joining. Let's start with the launch date.
Design is ready. We need two more days for testing.
Then we ship on Thursday and email customers on Friday.
Summary
- Launch moves to Thursday
- Testing needs two more days
- Customer email goes out on Friday
Used by teams worldwide
- Yale University
- New York University
Get more from every recording
99+ languages
Transcribe in the language that was spoken, then translate the transcript if you need to.
Speaker labels
See who said what. On every file with a paid plan or pass.
Language detected for you
Leave the language on automatic, or choose it yourself before you upload.
Long recordings
With the Weekly Pass or a paid plan, a 3-hour recording goes through as one file, up to 5 GB.
Meeting notetaker
Send a notetaker to Zoom, Google Meet or Microsoft Teams. The transcript is waiting when the call ends.
Custom vocabulary
Add names and terms once, so they are spelled right in your transcripts.
Made for the recordings your work depends on
Accuracy, measured
Word accuracy of our premium engine on a public test set of read speech, 80 clips per language, measured on October 20, 2026.
- English95.2%
- Spanish97.8%
- French97.2%
- German97.1%
- Portuguese97.2%
- Arabic82.9%
Word accuracy is 100 minus the word error rate. Clear audio scores higher than noisy audio, so names and numbers are worth a quick check.
Ask your transcript anything
Skip the scrolling. Ask a question and the answer comes from what was said in the recording.
AI chat comes with your first file. The Weekly Pass or Pro brings it to every transcript.
Connect your AI assistantRecord on your phone
Catt for iPhone and Android
Talk and the words appear as you speak. Bring in a recording, a video or a link, and find any sentence later. The app uses the same account as the website.
Record in your browser
Press Record live on this page and the words appear as you talk. Nothing to install.
Works with the tools you already use
- Google Calendar
- Zoom
- Google Meet
- Microsoft Teams
- Slack
- Public API
- Webhooks
Your recordings stay yours
- Encrypted in transitUploads and downloads travel over TLS 1.2 or newer.
- Encrypted at restStored files are encrypted with AES-256.
- No human reads your filesRecordings are processed by automated systems only.
- Not used for trainingWe never train on your content.
- Delete when you wantDeleted files leave the Trash after 30 days, or at once with Delete forever.
- Private by defaultOnly you can open your transcripts, unless you share a link.
For teams and business
For companies, agencies and teams. Priced around your volume and the way you work.
- Everything in Pro, for your whole team
- API access and webhooks
- Volume-based pricing for high call and recording throughput
- Dedicated support and onboarding
Frequently asked questions
How do I convert audio to text?
Upload your audio or video file on this page, or paste a YouTube or Vimeo link, then pick the spoken language or leave it on auto-detect. The transcript appears on screen when it is ready, and the first 30 minutes of any file are free with no signup. Sign up free to copy it or download TXT; Word, PDF, SRT, VTT and CSV come with your first file, the Weekly Pass or a paid plan.
Do I need to create an account to use the tool?
No. You can start transcribing immediately without creating an account. Sign up only if you want to save transcripts, access additional features, or manage multiple files.
What file formats are supported?
We support all major audio and video formats including MP3, WAV, MP4, MOV, M4A, WebM, and more. You can also paste URLs from YouTube, Vimeo, and other platforms.
Is there a file size or length limit?
Anyone can transcribe files up to 30 minutes free, with no account; the first file usually runs on our premium engine, later ones on our standard engine. A free account keeps that going: up to 30 files a day, up to 30 minutes each, and your first file can run up to an hour with an AI summary. A free account gets up to 2 hours a month at best accuracy, its first file included; after that, its files run at standard accuracy, without speaker labels, until the next month. If we ever hit our monthly capacity, free files may briefly run on our standard engine. Paid plans process the full file, and links and videos can run up to 24 hours each. On the free plan a link or video can be up to 3 hours long, and the per-file limit above still applies. A 3-hour lecture works in a single upload. Uploads can be up to 100 MB with no account, 1 GB on a free account, and 5 GB on a paid plan or with a pass or top-up.
Can I paste a YouTube or Vimeo URL to transcribe?
Yes. Just paste the video URL and we'll extract the audio and generate a transcript automatically. No need to download the video first.
How accurate is the transcription?
Clear audio on our premium engine usually needs only light edits, and you can fix any word in the editor while you listen. Accuracy depends on audio quality, background noise, and accents, so names and numbers are worth a quick check.
How can I improve transcription accuracy?
Use a clear recording with minimal background noise. Ensure speakers talk one at a time, speak clearly, and use a quality microphone. Better audio quality always means better transcription results.
Does it handle multiple speakers?
Yes. Every file on a paid plan or pass gets speaker labels, and a free account gets them on up to 2 hours a month. Our premium engine detects and labels the different speakers in your transcript, so conversations, interviews, and meetings are easy to follow.
Show 12 more questions
Does it work with accents and different dialects?
Yes. Our speech recognition engine is trained on a wide range of accents, dialects, and speaking styles to deliver accurate results regardless of how speakers sound.
Can I edit my transcript after it's generated?
Yes. You can edit any part of your transcript directly in the browser. Fix errors, add speaker names, insert notes, or highlight key sections before exporting.
What export formats are available?
You can download your transcript as TXT, DOCX, PDF, SRT, VTT, CSV or JSON. TXT is free on every account. DOCX, PDF, SRT, VTT, CSV and JSON need the Weekly Pass, a minutes top-up or a paid plan, except on your first file, which gets every format. Choose the format that fits your workflow, whether it's for documents, subtitles, or captions.
Can I add subtitles or captions to my video?
Yes. Once your transcript is generated, you can export it as an SRT or VTT subtitle file and add it to any video editor or upload it directly to platforms like YouTube.
Does it support translation to other languages?
Yes. Open a finished transcript in your dashboard and choose Translate to get it in any of 99+ languages, with each line's timestamp kept, then copy it or download TXT or SRT. Translating a finished transcript is included on paid plans, the Weekly Pass and the first file on a free account. Our Translate Audio and Translate Video tools also transcribe and translate a recording in one step, and Translate Subtitles translates an SRT or VTT file.
How long does transcription take?
Time depends on the length of the file: short clips come back quickly and long recordings take longer. The page shows progress while it works. On a signed-in account the transcript is saved to your dashboard, and we email you when a long file is done.
Can I transcribe in real time or only pre-recorded files?
Both. You can upload pre-recorded audio/video, paste a URL from YouTube/Vimeo, or hit Record live to transcribe in real time directly from your browser microphone. Recordings made without an account stop at 2 minutes. On a free account, a live recording uses your premium minutes while you have them; after that, each recording counts as one of your files for the day and stops at 30 minutes, or at 2 minutes if we reach our monthly capacity. On a paid plan or pass, a live recording uses the transcription minutes on your account.
What languages are supported?
We support transcription in 99 languages, including English, Spanish, French, Arabic, Hindi, Portuguese, German, and many more.
Is my data secure? Who can access my files?
Your files are encrypted during upload and processing. We do not share or sell your data. Only you can access your transcripts. Deleting a file moves it to the Trash, where you can restore it for 30 days. After that it is deleted permanently, and Delete forever erases it from the Trash at once.
Can I generate a summary of a transcript?
Absolutely! Our AI can generate smart summaries, key highlights, and action items from your transcripts automatically. Perfect for meeting notes and research.
Is ConvertAudioToText free to use?
Yes! Paid plans and passes run every file on our premium engine with speaker labels, and a free account gets it on up to 2 hours a month. A free account transcribes up to 30 files a day, up to 30 minutes each, no credit card, with a free TXT download, and your first file can run up to an hour with an AI summary. Anyone can also transcribe the first 30 minutes of up to 3 files a day with no account at all. Our free tools are completely unlimited with no account required. For more features, the Weekly Pass is $9.99 for 20 hours over 7 days, and Pro is $14.99/month, or $99.99/year.Yes! Paid plans and passes run every file on our premium engine with speaker labels, and a free account gets it on up to 2 hours a month. A free account transcribes up to 30 files a day, up to 30 minutes each, no credit card, with a free TXT download, and your first file can run up to an hour with an AI summary. Anyone can also transcribe the first 30 minutes of up to 3 files a day with no account at all. Our free tools are completely unlimited with no account required. For more features, the Weekly Pass is $9.99 for 20 hours over 7 days, and Pro is $19.99/month, or $119.99/year.
Can I generate transcripts from any video?
Yes! You can generate transcripts from any publicly available YouTube video, as well as upload your own audio and video files in MP3, WAV, MP4, and other popular formats.
Still have questions? Paid plans include help from a real person. See plans
Turn your next recording into text
Upload a file or paste a link. Your transcript opens as soon as it is ready.