Deriving Language names?

Help
1

Hi,

I’ve literally only just found out about the derive track language feature. I’m just wondering if anyone can help me with this track - 0002.mpls_3ita.flac what would i have to enter in terms of getting the program to identify italian?

2

Welcome!

The GUI works like this:

  1. First it splits the source file name into words by certain boundary characters, meaning everywhere one of those characters appears, the GUI will end the previous word & start a new word.
  2. Next it’ll iterate through the list of words from step 1 & compare them to the various forms of all the languages that should be used for language recognition. If a hit is found, that language is used.

Both the list of boundary characters as well as the list of recognized languages can be customized in the preferences → “Multiplexer” → “Deriving track languages”.

In your example the ita is surrounded by 3 & ., the latter of which is already part of the default set of boundary characters. However, digits aren’t part of the default set. Therefore you’ll have to add them to said list, e.g.

[](){}.+-=#0123456789

Then make sure that the right pane of “Recognized languages” (the one labeled “Selected”) contains Italian.

3

Hi, thankyou so much for the detailed reply, i totally understand how it all works. I have one problem, here is my track names - 00029.m2ts_ - 2 - DTS Master Audio, Japanese, 5.1 channels, 24 bits, 48kHz & 00029.m2ts_3jap.flac when i enter the above characters the languages are still in english. I believe the problem is that japanese is set to ja,jpn within the program and not japanese, jap as my files are showing. Would there be a way to solve the problem? Thankyou

4

Unfortunately jap won’t work, as you’ve guessed yourself already. MKVToolNix (and the whole Matroska file format) uses the family of ISO 639 language codes. MKVToolNix GUI compares the words from the file name to the two- & three-letter codes of a language as well as to its official English name (all according to the ISO 639 standard).

As you can see in the source code, Japanese (English name), ja (two-letter code) or jpn (three-letter code) would be recognized.

There’s no way to customize those codes.

5

Once again, thankyou for the quick response. These are the file names generated by eac3to -demux option. I dont know why the names come out that way. I will just have to name the languages manually, its not a big problem. Thankyou for your help, its much appreciated.

6

Yeah, these things (detecting languages or the audio delay from file names) are merely heuristics that work in a lot of cases, but certainly not all of them. That’s simply the limitation of not having proper metadata (e.g. a “track language” header field in the file’s format — Matroska has those, of course, but a lot of others such as MP4 or MPLS playlists do, too, but the elementary streams just don’t).