SaveForMeNow
EN
Remove background
ImageCompress imageResize imageCrop imageSocial media resizerInstagram without croppingConvert imageHEIC to JPGAVIF to JPGUpscale imageAI photo enhancerDPI checkerPassport photo makerRemove watermarkWatermark imageRotate & flipRemove photo locationSee all image tools →
PDFMerge PDFSplit PDFOrganise PDFCompress PDFEdit PDFSign PDFAdd page numbersWatermark PDFProtect PDFUnlock PDFJPG to PDFPDF to WordWord to PDFPDF to ExcelExcel to PDFPDF to JPGSee all PDF tools →
Video & AudioCompress videoTrim videoConvert videoMerge videosVideo to GIFRotate videoCrop videoChange speedMute videoVideo to textScreen recorderVideo to MP3Trim audioRingtone makerMerge audioVoice recorderSee all video and audio tools →
MoreAdd text to imageMeme makerCollage makerFavicon generatorWord counterCase converterPassword generatorJSON formatterPercentage calculatorUnit converterTime zone converterQR code scannerQR code generatorSee all design tools →
GuidesVideo downloader

Transcribe audio to text

Turn voice notes, interviews, lectures and podcasts into text. Free, private, and in around 100 languages.

Loading tool…

How do I convert audio to text?

Add an audio file, such as an MP3, M4A, WAV or a voice note from WhatsApp or your phone's recorder, pick the language and press Transcribe. The text appears with paragraph breaks at longer pauses, ready to copy into notes, emails or documents.

The speech model (OpenAI's Whisper) runs in your browser, so interviews and private recordings stay on your device. It works with around 100 languages and can also translate the speech into English.

Next steps:Compress audioTrim audioConvert audioVideo to text

How it works

  1. Add an audio file

    MP3, M4A, WAV, OGG, voice notes and more.

  2. Choose the language

    Or let it detect it automatically.

  3. Copy the text

    Or download TXT and SRT files.

Questions

Is it free?

Yes. There are no minutes to buy and no sign-up, because the speech model runs on your own device.

Are my files uploaded?

No. The audio is turned into text in your browser. Only the speech model itself downloads the first time (about 40–80 MB), and then it's cached.

Which languages work?

Around 100 languages, including English, Spanish, Portuguese, French, German, Indonesian, Hindi, Arabic and Chinese. Pick the language for the best result, or let it detect it.

How accurate is it?

Clear speech with little background noise comes out very well. Choose Accurate for accents, several speakers or noisy recordings, and check names and numbers before using the text.

How do I add the subtitles to a video?

Download the SRT file and upload it with your video on YouTube, Facebook or LinkedIn, or load it in a video player such as VLC. Most editors, including CapCut and Premiere, import SRT too.

How long can the recording be?

Up to 60 minutes. Long files take a while on phones; a computer is much faster.

This page in other languages:EspañolPortuguêsFrançaisDeutschBahasa Indonesia