Closed Caption Extraction from MPEG Stream problem

[sambat]

The mpeg clip lists CC'c in MediaInfo (eia-608) and are seen when the graph is played.
The dump bin file contains something(?) but CCExtractor doesn't see any information and outputs an empty .srt file.
Any ideas?

[usually_quiet]

Try another version of CCExtractor. A new one was released recently. If that doesn't work, try an older one. If nothing works, contact the CCExtractor development team to report your problem.

[gll99]

If the bin (raw) file is > zero bytes then in ccextractor under input options select raw McPoodles. Under output1 select srt(subrip).

edit:
I use a similar graph but mine is during a tv capture and this works every time. I missed that you're using an mpg as input.

[Cornucopia]

Aren't you trying to send that to a file sink?

Scott

[sambat]

If the bin (raw) file is > zero bytes then in ccextractor under input options select raw McPoodles. Under output1 select srt(subrip).

edit:
I use a similar graph but mine is during a tv capture and this works every time. I missed that you're using an mpg as input.

The output and execution window is gibberish.

1
00:00:12,078 --> 00:00:18,484
CCññ
ññYEñAHñ, Cñ

2
00:00:30,497 --> 00:00:47,212
ñññññññññññCC
C
ññSOñRRñY ñTRñAPñPEñR ñCAñN'ñ2Cñ

3
00:00:47,414 --> 00:01:00,726
ñññññññññññññ2CCxñ
ññBUñ2CCxT ñHEñ'Sñ BñEIñNC
ññYOñU ñJOñKEñRSñ WñERñE ñSñ
ññSUñCKñINñG ñHIñS ñ2CCxSOñCKC

4
00:01:00,927 --> 00:01:10,235
ñññññññññ2CCxCCññ
ññI ñKNñOWñ, ñBUñT ñ2CCxTRñAPñ

5
00:01:10,437 --> 00:01:20,646
ññññññ2CCxCCññ
ññCOñMEñ OñN,ñ
ññSHñAKñE ñ2CCxñ

6
00:01:20,847 --> 00:01:27,252
ññññññññññ2CCxCCñ
ñ2CCxRñHEñREñ YñOUñ Gñ

7
00:01:27,454 --> 00:01:33,859
ñññññ2CCxCCññ
ñ2CCxTñHEñY,ñ WñHAñ

8
00:01:34,060 --> 00:01:40,866
ñññññ2CCxCCññ
ññI ñTRñIEñD ñTOñ GñETñ
ññA ñ2CCxCOñUPñLEñ ñ

[sambat]

Aren't you trying to send that to a file sink?

Scott

The dump filter collects and saves the output to the .bin file.

[Bjs]

ProjectX generally works well thou setting the right code page for cc is another ... where a known code page dose not exist in the drop down box simply input the code page number and proceed with demux process.

[usually_quiet]

ProjectX generally works well thou setting the right code page for cc is another ... where a known code page dose not exist in the drop down box simply input the code page number and proceed with demux process.

Code pages don't apply here. These are N. American closed captions, similar to closed captions found on DVDs for the N. American market, which are unlike teletext or DVB subtitles. This form of closed captioning is stored in the GOP user data of the MPEG-2 video stream. ProjectX doesn't seem to work for that.

[Cornucopia]

CC is binary code that does not translate to simple ascii/ansi text. The MPEG userdata is, if IIRC, just a digitized version of the Line21 VBI analog data signal.

Scott

[sambat]

CC is binary code that does not translate to simple ascii/ansi text. The MPEG userdata is, if IIRC, just a digitized version of the Line21 VBI analog data signal.

Scott

Bugger.Two days chasing my own tail.
I was going to use the graph for an automated bat file for MPEG recordings, but in my ignorance I didn't notice that CCExtractor has a command line that I could use instead.
Sorry to all who responded for the time wasted.

[usually_quiet]

CC is binary code that does not translate to simple ascii/ansi text. The MPEG userdata is, if IIRC, just a digitized version of the Line21 VBI analog data signal.

Scott

Yes, that is right. The closed caption data is put on line 21 of the VBI when the file is played, as was true for the closed captioning technology used by the old NTSC analog broadcast system.