MP3 to text converter for MP3, WAV, M4A and MP4 files

sAInote, an AI voice-to-text app, converts an MP3 file into text directly in your browser, with no format conversion step first. It also accepts WAV, M4A, MP4, WebM, OGG and Opus, so the audio track of an MP4 video file works the same way an audio file does. Each upload can be up to 64 MB. Recordings up to 20 minutes are processed immediately on every plan, including without an account; recordings from 20 minutes to 2 hours run in the background and require Pro. Two hours is the hard cap per recording.

Convert an MP3 now 15 free minutes without an account, one time. Nothing to install.

What this converter handles

Seven formats, no conversion: WAV, MP3, M4A, MP4, WebM, OGG and Opus upload as they are. No re-encoding step on your side and no unsupported-format detour — hand over the file your recorder or phone produced.
Video files work too: MP4 and WebM are accepted, so the audio track of a video becomes text the same way an MP3 does. Text output only: no video editing, no subtitle files, no timestamp export.
64 MB per file: Each upload can be up to 64 MB. Size follows bitrate as well as length, so a high-bitrate export can exceed the ceiling while the same recording encoded lower fits under it.
Verbatim or edited: Quick Note returns the exact words. Smart Summary removes fillers and adds headings, paragraphs and lists without changing the facts. Both work on every plan, including without an account.
Edited and translated in one pass: In Smart Summary mode, Edit & Translate produces the cleaned text in roughly 35 target languages, translating the edited version rather than the raw transcript. Your language choice is stored on your account.
Findable later: Signed-in users get searchable history across web, iPhone, iPad and Mac. Search matches both the transcript and the processed output, and translated notes carry a badge. Guests get no history.

How to convert an MP3 to text with sAInote

1

Open sAInote

Open sainote.com in any browser, or use the sAInote app for iPhone, iPad or Mac. No account is needed to try it, and there is nothing to install for the web app.
2

Upload your file

Upload your MP3, WAV, M4A, MP4, WebM, OGG or Opus file, up to 64 MB. Recordings up to 20 minutes process immediately; 20 minutes to 2 hours needs Pro.
3

Choose how it is written up

Pick Quick Note for a verbatim transcript or Smart Summary for an edited version with headings and lists. Pro accounts can apply a saved Pro Assistant instruction instead.
4

Read, copy or keep it

Read the result and copy it out. Sign in if you want it kept in searchable history across your devices — guest conversions are not saved anywhere you can return to.

Which files this MP3 to text converter accepts

sAInote accepts WAV, MP3, M4A, MP4, WebM, OGG and Opus. You upload the file as it came off your recorder, phone or camera. There is no separate conversion step and no need to re-export into a supported format first.

MP4 is on that list on purpose. A video file carries an audio track, and that track is what gets transcribed, so a talk you filmed, a screen recording or a clip a colleague sent you can be turned into text exactly the way an MP3 can. What comes back is text — a transcript, or an edited version of it. sAInote does not edit video, generate subtitle files or export timestamps.

The format list covers most of what people actually have. M4A is what iPhone voice memos produce, OGG and Opus cover a lot of messaging-app voice notes, WebM is what many in-browser recorders write, and WAV is the usual output of desktop recording software. If a file falls outside the accepted set, converting it locally to MP3 or WAV first is the fix — sAInote does not convert file formats for you.

The 64 MB ceiling, and why bitrate matters more than length

The 64 MB per-file limit is the one people actually run into, and it is not a length limit in disguise. File size depends on bitrate — how much data per second the file was encoded at — as well as on duration. Two recordings of identical length can differ enormously in size because one was exported at a high bitrate and the other at a low one.

That works in your favour. A long interview exported at a high bitrate may come out above 64 MB, while the same recording re-exported at a lower bitrate fits comfortably. Uncompressed WAV is the format most likely to push you over the ceiling; compressed formats such as MP3, M4A and Opus are far more economical for the same material. Speech does not need a music-grade bitrate to be transcribed well.

So if an upload is rejected on size, re-export it at a lower bitrate, or in a compressed format, and try again. Splitting the recording into shorter pieces also works, and is the practical answer when the source file cannot be re-encoded.

Length limits: 20 minutes, 2 hours, and what happens in between

Two numbers govern length. Anything up to 20 minutes is processed immediately, on every plan, including as a guest with no account. From 20 minutes up to 2 hours, processing moves to the background and becomes a Pro feature: you start it, close the tab or the app, and come back to a finished note. Anything longer than 2 hours is rejected. That is a hard cap per recording, not a soft target.

Long recordings are handled differently rather than simply taking longer. Speakers are separated, and the recording is summarised hierarchically in chunks, so material from the last twenty minutes of a two-hour meeting survives in the output as well as material from the first twenty.

Usage is measured in minutes of audio, not in files. A 3-minute MP3 costs 3 minutes whichever mode you choose. Guests get 15 minutes, once. A free account also gets 15 minutes, once — a one-time allowance, not a monthly one, though the account keeps your history. Pro includes 360 minutes, six hours, every month.

Verbatim transcript, edited summary, or your own instruction

An MP3 can come back as one of three things. Quick Note is verbatim: what was said, nothing added and nothing removed. Smart Summary is edited — fillers dropped, broken sentences repaired, headings, paragraphs and lists added — while the facts stay exactly as spoken. Both are available on every plan, including to guests.

Pro Assistant is the third mode and is Pro only. You write your own instruction, save it, and apply it to any recording afterwards; a saved instruction is reusable and syncs to your account. If writing one from scratch is awkward, you can describe out loud what you want the assistant to do and sAInote drafts the instruction from that description. Each successful generation deducts 5 minutes from your balance, whether or not you keep the result.

Edit & Translate works in Smart Summary mode only. It edits and translates in a single pass into roughly 35 target languages, working from the cleaned output rather than the raw transcript, and it remembers your chosen language on your account so it follows you between the web app, iPhone, iPad and Mac.

MP3 to text: common questions

Can I convert an MP3 to text for free?

Yes. Without an account you get 15 minutes of audio, once, in the browser, covering Quick Note and Smart Summary. A free account also gets 15 minutes, once — that allowance is one-time, not monthly, though the account keeps your notes in searchable history. Pro includes 360 minutes every month.

What audio file formats can I upload?

Seven: WAV, MP3, M4A, MP4, WebM, OGG and Opus. They upload as they are, with no conversion step on your side. Anything outside that list has to be converted locally into one of those formats first — sAInote does not convert file formats for you.

Can I convert a video file to text?

Yes, if it is an MP4 or WebM file. The audio track is what gets transcribed, so a filmed talk or a screen recording becomes text the same way an MP3 does. The output is text only: there is no subtitle file, no timestamp export and no video editing.

What is the maximum file size?

64 MB per file. That ceiling applies to every format and every plan. It is separate from the length limits: a recording can be well under two hours and still exceed 64 MB if it was exported at a high bitrate, and lowering the bitrate is usually enough to fit.

My MP3 is larger than 64 MB. What can I do?

Re-export it at a lower bitrate, or split it into shorter files. Size depends on how much data per second the recording was encoded at, not on duration alone, so the same material at a lower bitrate can drop well under the limit. Uncompressed WAV is the usual culprit.

How long can a recording be?

Two hours, and that is a hard cap — anything longer is rejected. Up to 20 minutes is processed immediately on any plan. From 20 minutes to 2 hours, processing runs in the background and requires Pro, so you can close the app and come back to a finished note.

Can I upload several MP3 files at once?

No. Files are handled one at a time; there is no batch or bulk upload queue. For a set of recordings, upload them individually. Usage is counted in minutes of audio rather than in files, so ten short clips cost the same as one recording of the same total length.

Do I get timestamps or an SRT subtitle file?

No. sAInote returns text — a verbatim transcript in Quick Note, or an edited version with headings, paragraphs and lists in Smart Summary. There is no timestamp export, no SRT or VTT subtitle file and no caption track for video. Subtitle work needs a separate tool.

Can the text come back in another language?

Yes, in Smart Summary mode. Edit & Translate edits and translates in one pass into roughly 35 target languages, working from the cleaned text rather than the raw transcript. The language you pick is stored on your account, so it applies across the web app, iPhone, iPad and Mac.

Start with one file

Upload an MP3 — or a WAV, M4A, MP4, WebM, OGG or Opus file — and read the text back. Fifteen minutes are free without an account, and nothing needs installing to try it.

Convert an MP3 now