[mythtv-users] Vedr: Mythccextractor does not create a .srt file

Trond Nystuen trond_nystuen at yahoo.no
Wed Mar 20 23:34:52 UTC 2013


On 20/03/13 21:29, Jim Stichnoth wrote:

>> Does that mean the images are transferred in the stream, so there are no other way I can get the subtitles?
>
> That is quite possibly true.  Do you have any reason to believe that the stream contains any text subtitles, in addition to the 3 bitmap subtitle tracks?


I suspected mythccextractor created the images out of text subtitles, but I may be wrong. As you can see below, mediainfo says it is "DVB Subtitle", and the Codec ID is 6. I have not yet tried to find out what that means.


Anyway, I have not been able to get anything out of testing all teletext pages with ccextractor. So using the images seems to be the best (only) option right now.

The file names look like this:

0_52633905-to-52635705.png
1_52635825-to-52639705.png
2_52639825-to-52642185.png
3_52642985-to-52646265.png
4_52646425-to-52649145.png

I suppose the first part of the filename is a sequence number, and the two other ones are frame numbers.

Part of the information from mediainfo:

Text #2
ID                                       : 6510 (0x196E)
Menu ID                                  : 1 (0x1)
Format                                   : DVB Subtitle
Codec ID                                 : 6
Delay relative to video                  : 4s 374ms
Language                                 : Norwegian

Some tests tells me I am able to get the subtitles from the images with tesseract. I can probably convert frame numbers to timestamps for the .srt file. Mediainfo tells me the file has 25 fps. I may have to take the delay of 4s 374ms into account (or not).

This may work.

Trond



More information about the mythtv-users mailing list