Runs in your browser, no upload

Audio Compressor Smaller Audio, Nothing Uploaded

Choose a recording, pick Speech, Balanced or Music, and download a smaller MP3. Your browser does the work, so the audio never leaves your device and there is nothing to wait for but the conversion itself. Free, no account, no limit on the number of files.

Turn a recording into text

Free transcript with timestamps, in 50+ languages. Use the file you just made, or upload any audio or video file.

Drop your file or browse

MP3, M4A, WAV, OGG, FLAC, MP4, MOV, MKV, WEBM and more

Large video? Only the sound is needed for a transcript. Extract the audio first, or drop the video here and we will offer it.

10 free transcriptions, no account needed. Each recording can be up to 3 hours long, 5 hours in total, and a free preview of the transcript opens right here in your browser, with the full transcript in a free account. We run the models on our own servers, so a long recording can take a few minutes. If you create a free account later, everything you transcribed here moves into it, along with whatever is left of your free hours.

1

Choose the audio file

Drop an MP3, WAV, M4A, FLAC, OGG, AAC, WMA or another recording into the box above. The engine, about 30 MB, is downloaded once on first use.

2

Pick a setting

Speech gives 64 kbps mono, the right choice for lectures, interviews and voice memos. Balanced is 96 kbps stereo, Music 128 kbps stereo.

3

Download the MP3

You see the old and the new size side by side, can listen to the result, and save it with an ordinary download.

Why a voice recording does not need to be large

A WAV file stores every sample in full: an hour of CD-quality stereo is over 600 MB. Dictaphones, podcast recorders and some phone apps save exactly that, and even an MP3 exported at 320 kbps is about 140 MB per hour. For music that headroom can make sense. For one person speaking it is almost all waste, because speech uses a narrow band of frequencies and one channel carries it completely.

The Speech setting reflects that: 64 kbps, mono. An hour of lecture becomes roughly 29 MB and remains perfectly clear, small enough for email, messengers and course platforms with strict upload limits. Balanced keeps stereo at 96 kbps for interviews recorded with two microphones or podcasts with a music bed; Music at 128 kbps is the floor for songs, where lower rates start to sound thin.

Processed on your device

The compressor is ffmpeg compiled to WebAssembly, running inside this page. The recording is read from your disk and encoded in memory, and the MP3 is handed back as a download. Nothing is uploaded, which matters for interviews, medical dictation, client calls and anything else that should not end up on a stranger's server. Audio is much quicker to encode than video, even in a browser: an hour of recording usually takes a few minutes on a computer.

When compression will not help

A file that is already a small MP3 or a voice message from a messenger cannot be made much smaller without audible loss, and compressing it again only adds artefacts. The tool shows the saving after each run and tells you when there was little to gain, so you can keep the original. If the result is still too large, start again from the original with the Speech setting rather than compressing the compressed copy.

Compressing a recording for a transcript

It works well, and on a slow connection it is worth it. Speech recognition models convert every recording to 16 kHz mono before they listen, so a 64 kbps mono MP3 from the Speech setting holds everything the model uses: the transcript of a clear recording comes out the same as from the WAV, and a 600 MB file becomes about 29 MB to upload. Two things do make it worse: very low bitrates, below about 32 kbps, and compressing a file that was already compressed several times.

Transcrify also accepts the original, WAV and FLAC included, and uploads it in parts, so on a fast connection you can simply send that. 10 transcriptions of up to 3 hours each are free without an account, and a free account keeps the full transcripts. If you have compressed the file already, press Transcribe this file under the result and it goes straight into the transcription box below.

Frequently asked questions

Is my audio uploaded anywhere?

No. The compression runs in your browser and the file stays on your device. Only the engine is downloaded from our site the first time you use it.

Which setting should I use for a lecture or an interview?

Speech: 64 kbps mono. Voices stay clear and the file is small, roughly 29 MB per hour. Use Balanced if the recording is stereo on purpose, for example two speakers on separate channels.

Can I compress a WAV or FLAC file?

Yes. These are the files that shrink the most, often to a tenth of their size or less. The result is an MP3, which plays everywhere.

Can it make a video file smaller?

This page keeps only the sound. To keep the picture and reduce the size, use the video compressor.

Will compression make a transcript worse?

Not at the Speech setting. The recognition model works on 16 kHz mono audio, and 64 kbps mono MP3 keeps all of that, so a clear recording gives the same transcript. Avoid going much lower and avoid compressing an already compressed file again.

Compress your audio file

Scroll up, choose the recording and pick a setting. It is compressed in your browser and never uploaded.