Audio to text free, 10 recordings a day, video too

Plain transcript and subtitles right here. Bleeping, translation and burned-in subtitles live in the full tool.

How to convert audio to text without installing anything

1. Drop the file on the box above. MP3, WAV, M4A, FLAC, OGG, AAC and Opus all work, and so do video files if you only need the spoken words. Transcribe audio file to text straight from the recorder — there is nothing to convert first.
2. Wait about two minutes per hour of audio. The audio file to text transcription runs on our own hardware, not on your laptop, so nothing heats up and nothing installs.
3. Download the result. TXT for reading, SRT or VTT if you want subtitles, or copy the text straight off the page.

That is all of it. There is no transcription software to install and no sign-up. If you were about to download an app that transcribes audio on your phone, this page does the same job in the browser, and most of those apps want an account before they start. How to convert an audio file to text is the same three steps for every format, and that is also how to transcribe audio to text from a video or a phone recording. If you were wondering how to convert audio file in text form for notes or quotes, nothing changes, and the steps work the same when you convert audio mp3 to text. To transcribe recorded audio to text from a meeting or an interview, export the recording first and upload that file. To transcribe from audio that's on YouTube, use the YouTube page instead. When several people talk, the transcript already has speaker labels in it, and the subtitle files come along without a separate export.

Turning a voice recording into text takes about two minutes per hour of audio, whether it's a phone memo or a full interview.

Free transcription with no watermark, and how larger files are handled

Transcribing audio to text shouldn't come with a catch. Most "free" audio transcribers cap you at a few minutes a day or add a watermark until you pay. ScribeGrab runs on our own GPU rather than rented cloud time, so we don't pay a provider per minute — which means we can be generous: 10 recordings a day, individual files up to 60 minutes and 1.5 GB, around 100 languages auto-detected, and your file deleted the moment it's done. This box takes audio and video up to 1.5 GB and 60 minutes directly, splitting a big upload into pieces automatically, so a large audio to text job — a long lecture, a big WAV export, hours of call recordings — goes through right here in one upload. The homepage's full tool takes the same files and adds bleeping, translation and burned-in subtitles.

Interviews, lectures, podcasts and voice memos

Most jobs upload for one of a handful of reasons, and all of them follow the same three steps — get the file, upload it, download the TXT. Interviews and meetings become searchable, quotable text: record the call or save it as a WAV or MP3 from a handheld recorder, and pull the full record and direct quotes back out. Lectures and class recordings turn an hour of audio into notes you can skim and search, so you read the material once instead of scrubbing back through the recording. Podcasts come back as full transcripts useful for show notes and SEO: save the episode as an MP3 from your podcast app or the show's site, then transcribe it for a blog post. Voice memos — the note you dictated to yourself on your phone as M4A — come back as clean text you can read instead of replaying the audio. For a meeting recording, export the audio first: Zoom, Teams and Meet all let you download the recording after the call, and since ScribeGrab can't join a live call, grab that file first and transcribe it the same way as everything else.

MP3, WAV, M4A and other formats — and when the format matters

ScribeGrab reads MP3, WAV, M4A, FLAC, OGG, AAC and Opus files directly, plus common video containers when you only need the spoken words. There is nothing to convert first, so to convert sound file to text you drop it in as it is and copy the TXT back out. An MP3 to text job, such as a downloaded podcast episode, goes the same way.

A voice memo to text conversion, the note you dictated to yourself on your phone, works the same too, so the site doubles as a voice to text converter for dictated notes, and it's voice to text free within the same daily limit. If you came looking for voice to text online free as live dictation, note that this works from a recording you already have, not from your microphone.

Format barely changes the result: ScribeGrab decodes MP3, WAV, M4A, FLAC, AAC, OGG, Opus and AIFF the same way, so the words you get back don't change with the container — what differs is file size and where the recording came from. MP3 is the everyday podcast and download format: compressed and lossy, roughly 1 MB per minute, so a full episode uploads quickly. M4A is what iPhones and the Voice Memos app record in, and what iTunes and Apple Music use: portable and small. WAV is uncompressed studio quality, about 10 MB per minute, common straight out of a recorder or DAW. FLAC is lossless like WAV but compressed, the middle ground between full quality and a smaller file. You never need to convert one format into another before uploading: higher-quality source audio can help a little on borderline recordings, but a plain phone MP3 of clear speech transcribes just as cleanly as a studio WAV.

Coming from Evernote's built-in audio transcription, Eleven Labs, Kapwing, Clideo or Happy Scribe? It's the same basic job — upload, get text back — without the note-taking app, the editor or the account most of those ask for first. Speech to text and voice to text use this same converter, with tips for those kinds of recordings. Since every upload also produces subtitle files, see SRT vs VTT if you're not sure which one you need. Curious how close the transcript will be to what was actually said? See how accurate AI transcription really is for an honest breakdown. A Zoom recording is just another video file here: drop the .mp4 or .m4a Zoom saves and it transcribes the same way, or see transcribing a Zoom meeting for the speaker-label version. Speaking a language other than English? The transcript comes back in the language spoken — Spanish audio makes Spanish text, for instance — and translate subtitles takes it into English afterwards.

What transcribes best, and when to give it a proofread

Being honest about this saves you a surprise. Our own speech model is genuinely good, but recording conditions still matter:

• Near-perfect: one clear speaker, close to the mic, little background noise — a solo podcast, a dictated memo, a clean lecture.
• Very good with a quick proofread: two or three speakers taking turns, decent audio. A Q&A interview lands well; watch names and technical terms.
• Harder: heavy crosstalk where people talk over each other, lots of background noise, or heavily accented speech in a poor recording. It still transcribes, but read it through before you rely on it.

Background music generally isn't a problem — the model focuses on speech and ignores most non-speech sound. For the full picture, including what "95% accurate" actually means in practice, see how accurate AI transcription really is.

What people are actually uploading

In the 40 days from 13 July to 21 August 2026, ScribeGrab measured 5,361 uploaded audio and video files end to end. The average ran 18.0 minutes; the longest was 137.4 minutes, well past the 60-minute limit for a single free file. Most jobs cluster in longer recordings, not quick clips: 1,063 files ran 34–68 minutes, 817 ran 17–34 minutes, and 571 ran 9–17 minutes. 73% of everything uploaded sat comfortably under the limit, but 407 files landed within 10% of it and 197 went over it outright. The single most common way people get stuck here isn't a bad recording or a wrong format — it's length: "Media is longer than 60 minutes — please trim" was the single most common error on the whole site, hit 155 times in the same 40 days.

What you get back

Every upload produces three files, not just one, so you're covered whether you want text or timecodes:

• TXT — a plain transcript to copy into notes, a blog post, an email or a document.
• SRT — timecoded subtitles that YouTube and most video editors recognise.
• VTT — the web-native subtitle format for HTML5 players.

If your recording is really a video and you want captions rather than a transcript, head to video subtitles; to pick between the two subtitle formats, see SRT vs VTT. Need the opposite direction — writing read out loud as natural speech? That's our text to speech.

What you get, and the limits

ScribeGrab is a free audio to text transcriber, so you can transcribe audio to text for free — here are its numbers before you upload anything. One file can run up to 60 minutes and 1.5 GB; a €5 ShinobiTools pass raises the length to 4 hours and works on our other tools too. You get 10 recordings a day either way, and it resets at midnight. Around 100 languages are detected automatically. Your upload is deleted the moment the transcript is made, and the result itself is wiped after 45 minutes — 24 hours if you have a pass, because that longer window is part of what the pass is for.

What it does not do: it cannot join a live call or listen to your microphone, it does not translate while it transcribes (that is translate subtitles), and it will not write a summary for you — you get what was actually said, word for word.

A transcription audio to text example

This is what comes back. Feed it a two-minute voice memo and the TXT has one sentence per line, punctuated, ready to paste: "Right, the meeting moved to Thursday. / Can you let Marta know, and bring the printed quote? / I'll handle the rest." When two or more voices are heard, each line starts with a label such as "Speaker 1:". The SRT holds those same words in numbered blocks with a start and end time on each, which is what a video player needs to show them as subtitles.

Free transcribe audio to text, and what the limit really is

A free audio to text converter usually has a catch somewhere: a handful of minutes, a locked export, a nag to upgrade. This page runs on our own graphics cards, so the free tier isn't a trial with a countdown. The limit is 10 recordings a day of up to 60 minutes each, with no watermark and no account, and the same box reads audio and video files alike, so it also works as a free video transcriber. It is speech to text free for everyday use, and the paid pass is only for longer files and extras.

A note on the word "converter". An audio converter normally changes one format into another, and a free online converter that turns MP3 into WAV won't give you words. This one is an online audio to text converter: think of it as an audio converter to text online, where the output is a transcript. Each upload is a free convert to text, and used as an audio to text converter free of charge it has the same 10-a-day limit as everything else. Most people use it as an audio to text converter free online, from a phone or a laptop, and the audio to text converter online free tier has no install step. Some free converter sites stop after a few minutes; here a full hour gets transcribed in one go.

The same goes for the word "transcript". Used as an audio to transcript converter free of any sign-up, it hands you the words plus timed subtitles, so going from audio to transcript free of charge covers both. An audio transcribe to text job takes about two minutes per hour of recording, and you can transcribe audio free online, transcribe audio free from a phone, or audio transcribe free from any browser. This is audio text transcription, word for word, not a summary. It is an audio file to text converter, and audio file transcription is all it does, so free audio file transcription here means the full transcript, never a preview. Getting a transcript from audio file free of charge works the same for a one-minute memo and a 60-minute interview. Nobody needs to ask for text, subtitles or both: every upload gets all three files.

Compared with Word and Google Docs

Both have something for this, and both come with a catch. Microsoft Word audio to text is the Transcribe button in Word for the web (people also call it MS Word audio to text), and it asks you to sign in to Microsoft 365 first. Copilot's transcription lives inside the same Microsoft 365 apps. Google Docs audio to text is Voice typing, which writes down what your microphone hears live, so it cannot open a recording you already have. Google's Gemini can work with audio too, but inside a Google account and its usage limits. This page needs neither an account nor a microphone. It works from the file you already made, and an audio file to transcript is the whole job, with no extra step to extract anything first.

The same idea works on a scanned document instead of a recording: ScanReviver's OCR turns a scanned PDF into one you can search and select text in, instead of a flat image of a page.

Starting from a video file? Video to text takes MP4, MOV, MKV and WEBM directly.

By Sam Ridder — I build and run ScribeGrab on my own hardware, on my own. Who I am. Converting speech to text online gives you a file at a far smaller size than the recording it came from, which is why a two-hour interview fits in an email once it is text. You can read back the transcript alongside the audio to check names, numbers and spellings before you send it on.

Frequently asked