Here’s a problem from a complete novice. Exporting a speaking male voice recording project with a sampling rate of 96K to an mp3 streaming rate of 24K creates an artifact that is best described as an slight hollow sounding reverb on the voice. Here’s the setup: I have a audio recording made on a macbook using the iSight microphone (set to mono) and audiocity 1.3.7. The recordings are in a small carpeted office. The sample rate was 96K, 24bit. After recording, the noise is filtered out and the resultant signal is normalized. Playback on the project sounds just fine.
Now I export the project to mp3 at 24k streaming rate. Playback of the mp3 yields a very distinctive artifact that is akin to speaking in an office small office with no carpeting or drapes–like a slight hollow reverb to the voice. If I redo the export at say 32K the reverb artifact is reduced but not eliminated. Higher export streams further reduce the artifact. I need advice on knowing how to address this problem–assuming there is a fix, of course. The need for having a lower streaming rate is important.
32 is widely considered the lowest MP3 bitrate you can use to encode a mono, not stereo show with minimal damage. If the show is stereo it jumps to 64.
You are listening to the gargling and bubbling that happens to MP3 with too low a bitrate.
MP3 is rapidly falling out of favor and replaced by Apple AAC (iPods), Windows WMA, and I’m thinking one of the Flash variations of M4A or MP4 (YouTube). MP3’s only significant attribute right now is that everybody supports it. It was designed in 1987 and is showing its age.
Koz doesn’t mean that those other formats are not supported - he really means that MP3 has high portability among many diverse platforms. i.e. it will pay on a lot of different portable music players and computers/jukebox software.
AAC, for example is well supported by Apple Corp. (it is their proprietary format) - Microsoft supports WMA. But portability may be limited - AAC plays well on the iTunes s/w and the iPods it was designed for - but will not play on most “MP3” players - nor can you directly import an AAC file into Audacity. It is IMHO however a de facto global “standard”, purely down to the overwhelming dominance of iPods over other portable music players - and the cunningness of Apple in providing an iTunes version that works well on PC platforms ( a great marketing example of clever and slick vertical integration: mobile players(iPods)<=>multi-o/s platform management software (iTunes)<=>music via the iTunes Store).
MP3 is something of a violin, too. In the hands of a musician it can make grown men cry, but the “violin” setting on your keyboard might make a musician cry.
In the wild, lame has hundreds of controls, settings, and variations, only one of which is bitrate – although that is the most important. I wonder if you can get lame in the stand-alone form…
Nowhere is it written that you have to use lame inside Audacity, either. Won’t Switch or Super or other packages do this? I know you can sweet-talk iTunes Import Services to encode MP3.