Transcribe your audio recordings with ease.
or drop, paste, or choose from cloud
Auto-detect handles most recordings. You can also set the language yourself from more than 90 options. That is what saves a short or noisy clip from being read in the wrong language.
Download the transcript as a TXT file to paste into a document. Or take an SRT file with timestamps that a video player can load as subtitles.
The converter runs in your browser on any device. Transfers are encrypted and your recordings are never shared with anyone.
Upload your audio file.
Select the language spoken in your audio file.
Click the "START" button to start the conversion.
Download the text file.
Speech to Text is a free online tool that automatically converts spoken words from your audio recordings into written text. This feature can save you hours of manual transcription, making it useful for journalists, researchers, students, and business professionals. Whether you need to transcribe an interview, lecture, or meeting, our Speech to Text feature makes it quick and simple.
It is also called speech recognition, automatic transcription, or simply audio to text. A recognition model listens to the recording and splits it into small sound units. It then matches them against patterns learned from thousands of hours of recorded speech.
The result is a transcript you can read, search, quote, and edit. Nothing is typed out by hand, so a one hour interview takes minutes instead of an afternoon.
Accuracy depends most on the recording itself. Clear speech, one person talking at a time, and little background noise produce the best transcript. A phone recording made across a busy room will always be harder to read than a headset recording.
Setting the spoken language instead of leaving it on auto is the single easiest fix. Detection can guess wrong on a short clip, on a strong accent, or when 2 languages appear in the same file. One wrong guess affects every line that follows.
A processing speed setting controls the rest of the trade-off. The faster models return text sooner, while the slower models read the audio more carefully and get more words right. Pick a slower one when the recording is long, quiet, or technical.
Common audio formats work, including MP3, WAV, M4A, FLAC, OGG, and AAC. Batch conversion is supported, so you can add several recordings at once and get a separate transcript for each one.
Two output formats are available. TXT is plain text, ready to paste into a document or a notes app. SRT splits the same text into timed subtitle lines, which is the format video players and editors expect.
More than 90 spoken languages are covered. They include Arabic, Chinese, Dutch, French, German, Hindi, Italian, Japanese, Korean, Polish, Portuguese, Russian, Spanish, Turkish, Ukrainian, and Vietnamese.
Yes. Upload a recording, start the conversion, and download the transcript without paying and without creating an account. A subscription only comes into play for long recordings and for high daily volume.
It depends on where your audio is. Dictation apps type what you say into a microphone as you speak. This page does the other job: it takes a recording you already have and returns a text file. If the audio exists as a file, a converter like this one is the faster route.
Google Cloud Speech-to-Text is an API built for developers, and Google Docs offers voice typing for live dictation. Neither one lets you drop an existing audio file into a browser page and download a transcript. This converter does, with no code and no account.
Clear speech, a single speaker, and low background noise give the best results. Choosing the spoken language instead of auto-detect removes a common source of errors, and the slower processing speed settings read the audio more carefully. Expect to proofread names, jargon, and numbers.
More than 90, including Arabic, Chinese, Dutch, French, German, Hindi, Italian, Japanese, Korean, Polish, Portuguese, Russian, Spanish, Turkish, Ukrainian, and Vietnamese. Leave the setting on auto and the language is detected from the audio.
Common audio files work, including MP3, WAV, M4A, FLAC, OGG, and AAC. Several files can be uploaded in one go, and each one comes back as its own transcript.
Yes. Choose SRT as the output format and the transcript is returned as timed subtitle lines that a video player or editor can load directly. Choose TXT for a plain text file.
Yes. Open this page in a mobile browser and pick the recording from your phone or from cloud storage. There is no app to install, and the steps are the same as on a computer.
Everything here runs in the browser, with no install and no sign-up.
