An MP3 file is audio compressed with the lossy MP3 codec, usually from 128 kbps up to 320 kbps. It is small, plays almost everywhere and is the format most recorders and podcast editors export. Hushscript transcribes it as it is, with no conversion.

M4A is audio stored in an MP4-family container, most often compressed with AAC. Voice Memos on an iPhone is a common source. It is smaller than WAV at similar listening quality, and Hushscript takes it directly.

Which file format should I choose for transcription?

Use whatever your recorder or phone saved. If you can pick the setting before you record, WAV keeps the most detail and a normal MP3 or M4A saves space, and either is fine for speech. Do not convert afterward to improve the transcript, because a conversion cannot add back what was lost.

What is an OGG file and how do I open it?

OGG is an open container, most often holding Vorbis or Opus audio. Some players do not open it. You can upload an OGG file for transcription as it is, or convert it to MP3 with the free OGG to MP3 converter if a player or editor needs that.

What is the difference between an MP4 file and an MP4 format?

Nothing practical. MP4 is a container that can hold video, audio and metadata. The word format and the word file point at the same thing. What changes from one MP4 to the next is the codec inside it.

Can I transcribe a WebM or AVI file?

Usually yes. Hushscript reads the audio track in your browser and uploads only that audio. Whether a particular file opens depends on its codec and your browser. If one does not, convert it to MP4 with the free WebM to MP4 or AVI to MP4 tool and try again.

Do I need an AVI player to transcribe an AVI file?

No. Transcription does not play the video. The browser reads the file and takes the audio from it. A player only matters if you want to watch the AVI, and the free AVI to MP4 converter makes a copy that more devices open.

Should I convert my file before I transcribe it?

Usually not. Common audio uploads unchanged, and a compatible audio track is copied out of a video without re-encoding. Convert only when a file will not open, or when you want a smaller copy to share. Converting never adds detail the original lost.

Does file size change what I pay?

No. Minutes are counted against the length of the audio, not the size of the file. A 45-minute WAV and a 45-minute MP3 of the same recording use the same number of minutes.

What happens if Hushscript cannot read my file?

The job fails and nothing is charged. The app also tries once to convert the file in your browser and upload it again. If it still fails, convert the file to MP3 or WAV with the free audio converter, or write to support@hushscript.com.

Blog

Audio and Video File Formats for Transcription

Author: Hushscript Published: October 12, 2026 Last reviewed: October 12, 2026

Most audio and video file formats work for transcription. An MP3, M4A, WAV, OGG, MP4, WebM or AVI file all go through the same steps, and the format changes the result far less than the recording does. This guide explains what each audio and video file format is, which one to keep, and the few cases where converting first helps.

It covers the files you start with. For the files you get back, such as TXT, DOCX or SRT, see which transcript format you need. To transcribe a file now, use the audio to text page.

Format, container and codec: three words for one file

A file format is really two choices. The codec is how the sound or picture is compressed. The container is the file that holds the compressed streams together with metadata. MDN describes a container as a file format that encapsulates one or more media streams, such as audio or video, with metadata so they can be stored and played back together.

That is why two files with the same extension can behave differently. An MP4 can hold different codecs, and an OGG file can hold Vorbis or Opus. When a file will not open, the codec is the usual cause, not the extension.

Lossy and lossless audio

Lossless audio can be decoded to a signal identical to the original. Lossy audio throws some detail away to get a smaller file, and the more it is compressed, the more detail goes. MDN’s audio codec guide lists MP3, AAC, Opus and Vorbis as lossy, and FLAC as lossless.

Two rules follow from that. A lossy file cannot be turned back into a lossless one: saving an MP3 as WAV makes a bigger file with the same sound. And each lossy-to-lossy conversion costs a little more. Keep the original recording, and make copies from it.

The audio formats you will meet

MP3 file

An MP3 file is lossy audio, and it is the format you are most likely to receive. Recorders, podcast editors and download pages all produce it. At 128 kbps an hour of audio is about 58 MB, so it is easy to send and store. For speech, that is plenty. The step-by-step version is in how to convert an MP3 to text.

M4A file

An M4A file is audio in an MP4-family container, usually compressed with AAC. Phone recorders such as iPhone Voice Memos are a common source. It sounds about like an MP3 of the same size. See how to convert M4A to text for the phone workflow. If a player wants MP3 instead, the free M4A to MP3 converter does it in your browser.

WAV file

A WAV file usually holds uncompressed PCM audio, so it keeps everything the microphone captured and takes a lot of space. Stereo at CD quality (44.1 kHz, 16-bit) runs about 10.6 MB a minute, or roughly 635 MB an hour. Field recorders, studio software and some desktop recorders save WAV. Keep it if you have it. If you need to share it, WAV to MP3 makes a much smaller copy. More in how to convert a WAV file to text.

OGG and Opus files

An OGG file is an open container, usually holding Vorbis or Opus audio. MDN notes that an Ogg file with only an Opus track is sometimes called an Opus file and may even use the .opus extension, but it is still an Ogg file. Some players and editors do not open OGG. Hushscript accepts OGG, so you can upload it as it is. If another tool needs MP3, use OGG to MP3.

FLAC, AIFF and the rest

FLAC is lossless but compressed, so it is smaller than WAV. AIFF is uncompressed, like WAV, and common on Macs. Both are accepted. The upload picker also takes AAC, Opus, WMA, AMR and a few more. If you are unsure about a format, try it. Dropping the file in is the quickest test.

The video formats you will meet

For transcription, a video file is a carrier for its audio track. Hushscript reads the track in your browser, copies it out without re-encoding when it can, and uploads only that audio. The video never leaves your device. A large video does not make a large upload.

An MP4 file is the common case, a container many phones and cameras use for video. MP4 itself is only the wrapper, so what matters is the audio inside. Details in how to convert an MP4 to text.

MOV and MKV files work the same way. If another program needs MP4, MOV to MP4 and MKV to MP4 convert them in your browser.

A WebM file is a web video format, often from browser screen or camera recordings. Its audio is usually Vorbis or Opus. The free WebM to MP4 tool converts it for apps that want MP4.

AVI files are an older container from many cameras and programs. They often hold older codecs such as Xvid or DivX, which newer phones and editors may not open. Hushscript does not need to play the video to take the audio. If you want a copy that plays on more devices, AVI to MP4 re-encodes it at roughly real-time speed.

To save only the sound of a video, the free audio extractor keeps the original track where it can. The video to MP3 guide covers that task.

Which format gives the best transcript

The original one. Whatever your recorder or phone saved is the best file to upload, because it has been compressed the fewest times. Lossless formats such as WAV, FLAC and AIFF give the engine the cleanest signal. An MP3 at 128 kbps or higher of the same recording transcribes about the same.

The things that do move accuracy are outside the file: how close the microphone was, how much noise was in the room, and whether people talked over each other. Speaker labels, languages and exports do not depend on the format either. How to transcribe any audio file has the full method.

Length counts, size does not. Minutes are billed on the duration of the audio, so a large WAV costs the same as a small MP3 of the same minutes. A recording can run up to 10 hours with no fixed file-size limit, and large files are prepared in your browser, subject to your device’s memory.

When converting helps

A file will not open. Convert it to MP3 or WAV with the free audio converter, then upload that copy. Support depends on the codec and your browser, so a rare codec can still fail.
A tool or form wants a different format. MP4 for a video upload, MP3 for a player. The free converters run in your browser and do not upload anything.
You need to send a smaller file. Compress audio makes an MP3 at a bitrate you pick, and the original stays untouched.
You want one file from several. Join parts first with the free audio joiner.

When converting hurts

Do not convert to WAV or FLAC hoping for a better transcript. The detail that compression removed does not come back. And do not convert a file only because the extension looks unusual. Upload it first. If the transcription can read the file, nothing was needed.

If the server cannot read an upload, the job fails with nothing charged. The app then converts the file in your browser and uploads it once more on its own, so some odd files recover without any action from you.

Check the file before a long upload

Run the free 5-minute preview before you send a three-hour recording. It needs no account and shows how the text and speaker labels read on your own file. A format the preview cannot open on the page can still go through in the app.

The full transcript needs an account and enough minutes; a card is not required to sign up. Eligible personal accounts can receive 30 free minutes once, after card verification or with a first purchase. Pricing has the terms, and there is no subscription.

Independent sources and standards

Hushscript consulted these independent, non-competing references. They explain research, standards, or platform behavior and do not endorse Hushscript.

Sources reviewed: October 12, 2026