Free MKV transcription, no sign-up

MKV to Text Converter No Remux to MP4

Drop an MKV exactly as OBS or your screen recorder saved it, and the speech in it becomes text with a timestamp on every segment. 10 transcriptions are free, each up to 3 hours long. A free preview opens here without an account, and a free account keeps the full transcript.

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.

Need more than the free allowance? See the plans

1

Upload the MKV

Drag the file into the box above. It is sent in parts, so a multi-gigabyte stream recording does not have to be cut or compressed beforehand.

2

The audio track is decoded

The picture and any embedded subtitle tracks are skipped. The sound is decoded on our servers, the language is detected and each segment gets a timestamp.

3

Read the transcript

The text opens in your browser. Copy the parts you need, or create a free account to keep the full transcript and correct it in the editor.

Why so many recordings end up as MKV

OBS Studio records to MKV by default, and for a good reason: if the program crashes or the power goes out, a Matroska file is still readable up to the last second, whereas an MP4 that was never finalised is usually lost. Streamers, teachers who record lessons and anyone capturing a long meeting from the desktop therefore accumulate MKV files without having chosen the format.

The same container is what many screen recorders on Linux write, what video capture cards save, and what tools like MKVToolNix and HandBrake produce when several tracks need to live in one file. The downside shows up later. Editors, phones and a good number of web tools do not open MKV, and the standard advice is to remux to MP4 in OBS first. For transcription that step is unnecessary. The file is read here as it is.

Several audio tracks in one file

Matroska can hold many audio tracks side by side. OBS users often record the microphone on one track and game or desktop sound on another, and films carry a track per language. A transcript is made from one track only. When a file contains several, it is normally the first one, the track marked as default, and there is no option on this page to pick a different one.

So it pays to know your layout. In the usual OBS setup track 1 is the full mix, with voice and desktop sound together, and that is the one transcribed. If your microphone sits alone on track 2 and you want only the voice, extract that track with a tool such as MKVToolNix or ffmpeg and upload the audio file instead. Subtitle tracks stored in the MKV are not read; the text always comes from the speech itself.

Long streams and big files

A three-hour stream at a high bitrate can weigh twenty gigabytes. The upload is split into parts, so the size of the file is not what stops you; your connection decides how long the transfer takes. The limit that applies is duration: up to 3 hours per recording as a guest.

If the upload is the slow part, you can pull the sound out first. The extract audio and video to MP3 tools on this site run in your browser and turn the recording into a file a small fraction of the size, which then transcribes exactly the same. Game audio and music under the voice lower accuracy more than any codec does, so a recording where the microphone is clearly louder than the background reads best.

What is free and where the file goes

As a guest you read and copy a free preview of the transcript. A free account holds the full text and adds the editor and a history of your transcriptions. The free allowance is 5 hours across 10 recordings, shared between guest and account. Export as SRT, VTT or ASS subtitles, AI chat with the recording, summaries and PDF or Word export come with the paid plans.

Speech recognition runs on servers we operate, and the recording is not handed to a third-party speech API. For unreleased video, recorded client calls and internal training sessions, that is usually the question that has to be answered before the file can be uploaded anywhere.

Frequently asked questions

Do I have to remux the MKV to MP4 first?

No. Upload the MKV as it was recorded. Remuxing changes only the container, and the audio that gets transcribed stays the same.

My MKV has more than one audio track. Which one is transcribed?

One track, normally the first or default one. In a standard OBS setup that is the full mix. There is no track selector, so if you need a different track, extract it to an audio file and upload that.

Are the subtitles inside the MKV used?

No. Embedded subtitle tracks are ignored. The transcript is produced from the spoken audio, with its own timestamps.

Can I upload a very large OBS recording?

Yes. The file is uploaded in parts, so size alone is not a barrier. Length is what counts: up to 3 hours per recording as a guest, with 10 free transcriptions and 5 hours in total.

Can I turn the transcript into subtitles for the video?

The transcript is timestamped, and a paid plan exports it as SRT, VTT or ASS. A free account lets you keep and edit the full timed text.

Transcribe your MKV recording

Scroll up and drop the file in as OBS saved it. 10 transcriptions of up to 3 hours each are free, and the text is ready a few minutes after the upload finishes.