View Full Version : Incomplete splits

08-11-2008, 08:04 PM
When recording a Shoutcast broadcast as a stream with the "Split Shoutcast MP3 streams into individual files" option selected, is there a way to adjust the timing of the split that I might have missed?

I have found that the first part of the song is chopped off. You can find the missing piece at the end of previous song and piece it together using Audacity, but there is still about .150 seconds missing as a result of the split. This gap can be unnoticeable after the file is pieced back together if the audio is consistently the same, but if the music is changing pitch or if there is an obvious beat it can be noticeable no matter how much tinkering you do.

Second, I also selected "Keep Copy of Original File After Converting" when recording, but the .cap file still disappears after conversion. Why is that?

Lastly, I tired copying the .cap file off to a new location before the recording session ends and the .cap file is split and deleted thinking I could convert it as one big file and manually split out the tracks I could not fix. Unfortunately when I try to convert it with "Replay Converter" it isn't recognized with the .cap file extension. If I manually change the file extension to various file extensions such as .mp3, audio programs will play the file but there are gaps of silence scattered throughout. I guess my question is, what format is the .cap file and how do you convert it to a usable audio file?

I searched for answers in previous postings but came up empty. If the answers are available elsewhere, my apologies.

Cheryl Wester
08-12-2008, 05:21 PM
Go to the properties of the show you are capturing. Now go to the splitting tab and work on adjusting the time milleseconds-this does only work for audio capture and not stream capture, however.

A CAP file is a temporary file. When you are done recording it is going to change to the format that it originally recorded in. You should see both that file and the converted file format. If you have a cap file it normally means that the program shut down prior to the end of the recording and before it had time to convert to the native format. That could also happen if you have an Internet glitch.

08-12-2008, 09:30 PM
There are three problems recording a session by selecting “Capture by Recording Audio Output” instead of "Capture by Downloading Streams"

1. If you are capturing “audio output” that infers an analog signal. That means the signal arrives from the Internet in digital form that is converted by the PC sound card into an analog signal to provide audio output which is then reconverted to a digital signal so that Replay AV can store the information in digital form on the hard drive. If that is the case, there are two extra conversion steps that will degrade the audio from what is already a marginal feed from the Internet. It is like taking the audio output of a CD played through a sound card or amplifier and digitizing it rather then “ripping” the digital information directly from the CD. In the former you end up with all the distortions the intermediate hardware adds to the signal as well as the original signal.

2. When you select “Capture by Recording Audio Output” you lose the ability of the Replay A/V software to rename the files based on the information in the stream. Instead of file names that includes the artist, title, and any other information embedded in the stream, you end up with a series of files with the names similar to “New Show Aug 12 01” through “New Show Aug 12 n”. This means you have to rename all the files after the fact - and that is only if you are lucky enough to recognize the artist and know the title from listening to each file.

3. Time splitting is not perfect either. Even if you select the best possible time you can still end up with tracks split in multiple parts as if they were single tracks as well as having two tracks not split, all in the same session.

The best solution to my way of thinking is to be able to adjust the time based on the when the embedded information arrives, or ideally, a combination of factors such as dead space within so many milliseconds of the arrival of the embedded information.

Since based on your response that is not possible, the best solution is to continue recording the stream and letting the software split it into tracks so I have the artist and title information, and then trying to correct the problem of losing the first part of the track by using cut and paste from the non split original session file.

If I read your response correctly, the reason I do not have an “original file” after the conversion of the .cap file when the recording session ends is because the format of the split file (.mp3) is the same as the as the original (also .mp3). As a result I selected convert the output as an .ogg file under “Basic” as well as keep a copy of the original file under “Output”. This should mean that after the conversion at the end of the session, I should be left with .ogg files split into individual tracks and a large .mp3 file of the complete session. When I tried that I did not obtain the result I expected. It still only provided the session split into .mp3 files that need editing to add the missing front of the file and to delete the beginning of the next track at the end of the file. There was no complete session file and there were no .ogg files.

Since that approach did not provide one long session file, I am now back to trying to do something with the .cap file, as I described in my original posting, to generate a usable session file.

As you stated in your response, the .cap file is normally deleted after the conversion at the end of the session, and I understand that. What I did is create a batch file, that when used with the Task Scheduler, copies the .cap file to another directory a couple of minutes before the session ends. In that way I have copy of the captured stream as one long file as well as the split .mp3 files with the title and artist information even though the front is chopped off. What I need to do is convert that .cap to a file that Audacity can read.

In order to do that I am again asking what the format is for the .cap file?

If you have any other suggestions as a work around to the problem I am having other than what has been previously tried, I would be willing to try them.


08-21-2008, 04:31 AM
In order to do that I am again asking what the format is for the .cap file?

Can someone answer this? Thanks

Cheryl Wester
08-21-2008, 09:16 AM
A cap file is a temporary file. It is a file that did not finish closing correctly. To get the file to play back all you need to do is rename the extension what it shour of been originally and it should work for you.