MP3 to Text Converter
Turn an MP3 into an editable transcript for free, right in your browser. Your file never uploads, and you can download text or captions.
How to convert MP3 to text
- Drop your MP3 in the box above or choose it from your device.
- Choose Fast or Accurate, and set the language if it isn't English.
- Press Transcribe and watch the transcript fill in.
- Edit it, then copy it or download a .txt or .srt file.
Want to know how long the transcript runs when you read it aloud? Paste it into the words to minutes calculator.
MP3 transcription for podcasts, interviews and voiceovers
The MP3 stays in this browser tab. Script Timer does not receive the file, and closing the page leaves nothing on a server. The speech model downloads once from the Hugging Face hub and is cached for the next episode.
Most MP3 to transcript sites send your file to their servers and make you wait in line. This one doesn't, which makes it a good fit for podcast episodes before they're public, recorded calls and rough interviews.
Get a cleaner MP3 transcript
- Use the original export, not a re-compressed copy. Low bitrates blur consonants.
- Trim long music intros first. The model can try to transcribe lyrics.
- For two or more speakers, use Accurate and add speaker names as you edit.
- Split files longer than an hour into parts.
Turn the MP3 transcript into the next episode
Pull the best lines from the MP3 transcript for show notes, then paste the full text into the words to minutes calculator to see how long the episode runs.
Read the trimmed script on the free teleprompter, or hear it in 100+ AI voices before you record the next episode.
More free audio converters
MP3 to Text FAQ
Is my MP3 uploaded anywhere?
No. The MP3 is decoded and transcribed in your browser. Script Timer never receives the file. The speech model downloads once from the Hugging Face model hub and your browser caches it.
Is MP3 transcription free?
Yes. An MP3 transcript needs no sign-up and adds no watermark. One MP3 can run up to 60 minutes, which is as long as the browser can handle in a single pass.
What model reads an MP3?
OpenAI's open-source Whisper model, run in the browser with transformers.js. Fast uses Whisper Tiny, which is a good first pass on a clear podcast. Accurate uses Whisper Base when the MP3 has music beds or more than one speaker.
Why is the first MP3 slower?
The first MP3 waits while the browser downloads the speech model. The next episode uses the cached model and starts right away. After that, speed depends on your computer or phone.
Can I turn an MP3 podcast into show notes?
Yes. Transcribe the episode, copy the text, and pull your best lines and timestamps into show notes. The timestamped download makes it easy to find each moment.
Hear it before your audience does
Paste your script into Script Timer to time it, grade the delivery and hear it in 100+ AI voices.
In-browser speech recognition follows whisper-web (MIT license).