5 min read
MKV to MP3: extracting audio from Matroska files
Matroska files often hold several language tracks and a commentary. Extracting audio means choosing which stream you actually want.
MKV shows up in documentary and archive work far more than in narrative production. Screeners, festival copies, archival transfers and camera output from some encoders all use Matroska because it accepts almost any codec and any number of streams.
One container, several audio streams
A single MKV can carry an original language mix, a dubbed version, a descriptive audio track and a director commentary. Each stream has its own codec, channel count and language tag. Extracting audio without checking which stream you grabbed is how a French dub ends up on an English cut.
- Check the stream language tag before extracting.
- Note the channel count: a 5.1 stream extracted as stereo has been folded down by the tool.
- Note the codec: AC3, AAC, Opus, FLAC and PCM all appear in MKV.
MP3 or WAV
If the MKV audio is already lossy, extracting to WAV does not restore anything, it just makes a bigger file. WAV is still the right choice when you plan to edit, because every workstation handles it natively and nothing re-encodes on the way in.
Surround streams
A 5.1 stream needs a decision, not a default. Either keep all six channels for a proper conform, or fold to stereo with a documented LoRo downmix so the dialogue stays intelligible.