I have Win 7 professional and Audacity 2.2.2
I put together a radio show once a week for our local FM radio station. I make nine 15 minute segments, and they are a mix of my voice overs and songs I have saved as MP3’s on my computer. I am very careful to make each segment as close to 15 minutes total and usually come within a few seconds.
Recently when the segments are in their system and they put them into their computer the total time of my segments are not the same, they are several minutes longer that what was delivered.
I have been trying to help them figure out what the issue is and they are stumped too, so I thought I would ask here just in case anyone can help.
my settings are 44100hz and the bit rate I was using was 32 bit float (that setting I have changes to 16 bit)
one other thing, when i record my voice overs i have been doing them in mono, not sure that makes any difference but thought I’d say so.
So sometimes i may have 4, 5, 6, or more tracks to mix as one segment.
I am attaching a screen shot they sent me of what they are seeing on their end.
We have been doing this from Audacity for about 6 months and just recently have we been having problems. They use Adobe Audition but it has been working fine up until recently
Any Ideas what is happening?
It’s probably related to variable bitrate… I thought these problems were fixed years ago…
If you need MP3 try exporting as 320kbps CBR (constant bitrate). That’s the “best” MP3 setting and the constant bitrate makes it easier for the software to figure-out the length.
Or you can try “repairing” the MP3s with [u]VbrFix[/u] or [u]MP3 Diags[/u], but you probably don’t want to do that every time.
my settings are 44100hz and the bit rate I was using was 32 bit float (that setting I have changes to 16 bit)
The MP3 file doesn’t have a bit-depth. You should let Audacity “work at” 32-bit floating point which is the default.
Yes.
A few years ago, Audacity switched from “128 kbps CBR” as the default format for MP3, to “Preset Standard” as the default format.
(“CBR” is the abbreviation for “Constant Bit-Rate”).
“Preset Standard” uses “variable bit rate”, where the bit-rate increases for more “complex” (difficult to compress) sounds, and reduces the bit-rate where the audio is easier to compress. “Preset Standard” averages at around 200 kbps for stereo tracks, and somewhat less for mono tracks (whereas “128 kbps CBR” is 128 kbps regardless of the number of channels or complexity).
The reason for the change was that most Audacity users are looking to achieve reasonably good quality stereo music recordings, which is best achieved with “Preset Standard”. Preset Standard will also produce good quality voice recordings, and the same quality for mono or stereo.
Unfortunately, there is a common bug some software, which is that it reports the wrong length for VBR MP3. There really is little excuse for this bug as VBR encoding has been around for well over a decade, but sadly there’s still a lot of software that can’t be bothered to fix it.
Because of the prevalence of this bug, it is usually recommended to use CBR encoding for podcasts and similar. 128 kbps CBR is a good choice because it is properly supported by just about everything (though “Preset Standard” still gives better sound quality for stereo recordings).
See here for where to set “128 kbps CBR” (Constant bit-rate: 128 kbps): https://manual.audacityteam.org/man/mp3_export_options.html
However.
One of our posters was doing something very similar to what you’re doing, but he ran into other bitrate problems. He, too downloaded internet music or other MP3s and commented on it and he, too supplied the station with good quality MP3s. They broadcast the shows and everything was OK until the station posted the podcast version of the shows. That was yet another MP3 conversion and the voice survived, but the music turned into garbage.
See if they will allow you to submit in WAV (Microsoft) 16-bit instead of MP3. The files will be much larger, but they won’t have that slowly falling apart problem, and they won’t have the MP3 codec support problem, either.
Koz