Hi, using A2.0.5 win 7pro sp1, export to mp3 via lame, recording from old cassettes, spoken meetings, content valuable - sound quality less so - what settings can I use to reduce exported file size and still be as good a quality as original cassette-which is marginal to start with. I tried the first one, a 1.5hr meeting exported as 80Meg mp3 with defaults, using 64K output became 40Meg. Are there other settings during record, like sample rate or whatever that could reduce further? I can detect no diff between the 128k and 64k with my ear so again, not needing super audiophile output. Thanks
The main determinant of MP3 file size is the bit rate, so if 64 kbps is still too large, try reducing the bit rate further.
If the recording is stereo, use Tracks > Stereo Track to Mono to make it mono. This will let you reduce the bit rate further for the same quality than if you had a stereo track.
Set Device Toolbar to record in mono if you have not yet done so.
If you use variable bit rate MP3 export, you can reduce the Audacity project rate to reduce the exported file size a little more.
Thanks for the quick response. What about the section under Edit/preferences/quality/sample rate - sample type etc settings…would there be minimums i could set those to, which would reduce export size without damaging quality?
32 is the minimum MP3 quality for mono sound. I found that out by actual experimentation and then a bunch of my friends found the same number independently. Above 32, you can ignore the compression damage and enjoy the show if you never heard the original or had nothing to compare it to. Below that, everything turns into a bad cellphone call.
I listen to a broadcast radio show that produces a podcast with an internet introduction. The show quality is OK, but the introduction is badly compressed and sounds nine-year-old grade-school dreadful compared to the show. I see in the last couple of shows the quality of the intro is much better. I bet there were complaints about it.
Below 32, sound turns to bubbly, honky trash no matter what. There is a way to improve the compression a bit by using lame in the stand-alone manual mode and not internal to Audacity. I’ll see if I can find that posting.
Here it is. It was a discussion about doing almost what you want.
Those do not affect export size of MP3 with the exception of sample rate. Reducing sample rate will somewhat reduce MP3 file size if you use Variable Bit Rate MP3 encoding, but (generally speaking) not if you use Constant Bit Rate.
If you decide you like 22050 Hz Variable Bit Rate, you can set Default Sample Rate in the Quality Preferences to 22050 Hz then your recordings and exports will be at that rate. 22050 Hz is usually a good rate for high quality speech recordings. You could use a lower rate if there is not much treble in the speeches on the cassette.
You can look in Analyze > Plot Spectrum… at a recording you made at 44100 Hz. If it does not show higher than for example 5000 Hz, then an export sample rate of 11025 Hz would be OK. A given sample rate captures frequencies up to just less than half that rate, so 11025 Hz would contain up to about 5000 Hz comfortably.
Keeping in mind 5KHz sound is the quality of AM Broadcast.
Half of a 10 KHz AM channel.
So if you like talk radio, this is for you.
Fantastic! Gale and Koz thank you so much for that info, exactly what I was looking for! mil grazi