compress archive playback?

[jonwrice]

Is there any way to "compress" playback from the archives to reduce the dead air time?

Thanks,
Jon

 

[slash]

None that I know of...
In my opinion it would be really nice to have a spectrograph of the audio to accompany the web player when playing back of the 15-minute audio clips, so one can skip to where there is a spike of radio traffic in the timeline without losing your sense of response times.

 

[blantonl]

There actually is, and we're looking at how me might implement this on a wide scale basis.

 

[j2brown]

It may not be appropriate for use site wide, but I used to run the silencer program available here:

DNA Lounge: Source Code: MP3 Archives

I had a script on my Linux box that kicked off streamripper to record my local stream for a set period of time, then passed the output to silencer when it was done, then put it in an archive folder.

However, since the source code is available maybe someone in the RR talent pool can make it useful for the site as well.

jeff
sdg

 

[blantonl]

Jeff, nice catch. We've been looking at using the sox framework to accomplish this, however this might be an easier approach. Thanks!!

 

[GTR8000]

While we're on the topic of the archive files, were you ever able to figure out what Streamripper does with that 10 seconds or so between files that it seems to lose? Or is that just an unfortunate consequence of running so many instances of Streamripper that it lags a bit when starting up a new file?

I've used Streamripper for quite a few years and have never come across lost data like that. If anything, I experience the opposite effect andget a few seconds of duplicated data between files due to it picking up buffered audio at the beginning of each file. Then again, I'm not ripping hundreds of streams at once, so it's really unfair to compare the two situations.

Also, is changing to 30 minute files still in the plans as a method to reduce the lost data, or no? I was actually looking forward to a change to 30 minute files, as the 15 minute ones are a bit tedious to work with for long incidents, particularly having to download so many of them one at a time.

 

[hvscan]

I've used Audacity to remove the dead air time. It's free and easy to use.

 

[blantonl]

We're working on it Chauffer... be patient

 

[PeterGV]

There's some potential for RR to "value add" the audio archives. As you consider how to deal with long periods of silence, you also might want to simultaneously consider how you deal with storing tags as part of the audio archive stream (note that tags are presently not available in the archives, which limits the usefulness of the archives considerably for some feeds).

The existing MP3 files are kinda the least common denominator for audio playback and should be maintained. One way of handling the location of audio bits in a stream might be to create an index file containing time markers for non-silence periods with the associated tag. An RR Feed Aware player could then list the non-silence periods (with associated tags) and even play back (online) selected segments or allow them to be downloaded and stored (with a file name comprising the tag, date, and time).

I've been considering doing something similar for a ScannerCastRecorder...

Peter
K1PGV

 

[jasonk]

Please do not Change to 30 Minute Archives ......At least get some feedback before you did that ....There are plenty of Download Managers out there that make D/L a days worth of archives Simple. 15 Minute Files are perfect to work with in any Audio Editor .....Much larger Filesize or Time Delay between archives would really Decerase the enjoyment value of Archives.

Jason Kinley

While we're on the topic of the archive files, were you ever able to figure out what Streamripper does with that 10 seconds or so between files that it seems to lose? Or is that just an unfortunate consequence of running so many instances of Streamripper that it lags a bit when starting up a new file?

I've used Streamripper for quite a few years and have never come across lost data like that. If anything, I experience the opposite effect andget a few seconds of duplicated data between files due to it picking up buffered audio at the beginning of each file. Then again, I'm not ripping hundreds of streams at once, so it's really unfair to compare the two situations.

Also, is changing to 30 minute files still in the plans as a method to reduce the lost data, or no? I was actually looking forward to a change to 30 minute files, as the 15 minute ones are a bit tedious to work with for long incidents, particularly having to download so many of them one at a time.

 

[jland138]

Recording feed audio

I like the interface of Scanner Recorder Pro:

Scanner Recorder Pro

It has interval, event and elapsed time recording. Can split left and right channels into separate files and with different settings. Records time and tags. Can even split files on the fly. Player will display event ID and time of recording. Player also has the ability to skip events. More than worth the $10 donation.

I wish I could find something similar for non-Windows systems.

 

[blantonl]

Much larger Filesize or Time Delay between archives would really Decerase the enjoyment value of Archives.

Jason,

Thanks for your feedback.

Would you share with me your thoughts on how increasing the archives to 30 minutes would decrease the enjoyment value of the archives?

 

[GTR8000]

There are plenty of Download Managers out there that make D/L a days worth of archives Simple.

Can you please share what download manager and what settings you are using to download multiple archive files in one shot? I've tried Download Them All in Firefox and it never gets all the links. Thanks.

 

[vernondcole]

I would like to insert my thoughts on this subject. The stream I just started providing is for Fire and EMS in a small desert town. If there is 30 minutes total of actual traffic during a day, then it was a busy day.
The other 23 1/2 hours is a plain waste. What I really want is a "recall" button on my smart phone that will replay the last voice traffic on the air (and then the one before that) so that when my pager gets a garbled signal, I can use the scanner stream to get a clean copy. And I'ld really like it to work while I am running to my car, not sometime tomorrow.
I also do not want to eat up the battery and bandwidth on my Android by listening to the hum and hiss of an old Bearcat on squelch. There should be what they call in the recording industry a "noise gate" which, if the signal is close to zero for a given interval, makes the signal _exactly_ zero until the next time a significant signal comes in. My VOIP soft phone has the same feature, except there it is called "silence suppression."
If this is too hard to do on a server (and I can see that it could be) then perhaps the best answer is something like an "icecast_3" where the filtering is done by the sending computer, which has only one, or very few, streams to watch. I will send you a packet of sound only if there is something in it -- and a heartbeat every few minutes if there is not. I realize that this would be mean creating a new "fork" of icecast -- one optimized to burst transmission rather than entertainment broadcasting. That might not be a bad thing.