Hi, I recently acquired a cheap chinese MP3 Player with voice recording/encoding capability.

The max voice quality you can get from this is "IMA ADPCM 16bit 32kbps mono".

I downloaded some voice encodings from the player and opened them with Adobe Audition, edited useless parts and saved as the same format.

There is supposedly a re-encoding process going on, but I can´t notice any quality degradation. I really don´t want to save to plain PCM because, correct me if I´m wrong, the files would be 4 times bigger (I already tried with approximately 4x size increase).

What would you do to preserve as much quality as possible without increasing the size (I mean keeping it around the same bitrate)?

Thanks.

Edit: it was actually 32kbps

If all you're doing is splicing, you wouldn't lose any quality except at boundaries. I think you can safely ignore that.

I'm, making something similar to Megaman, but I can reach also 128kbps.

I don't have Audition. So I'm searching a free program that can cut IMA files without recompress them, like that excellent prog I use for cutting mp3: Mp3DirectCut.

Also I'm not sure I understand why you don't loose quality on recompress IMA while you will have if you recompress mp3 or other lossy codecs.

If you crop a .jpeg file, you'll have to recompress. You'll lose quality. On the other hand, if you crop a 255 colour .gif file, you still have to recompress, but you won't lose quality.

ADPCM is a bit like .gif; quantization will be lossy but the storage is lossless.

Previous thread with some more adpcm info.

jmvalin mentioned something about ADPCM on the ogg-dev mailing list at xiph.org.

On 11/14/05, he wrote:

<snip>

> but with ADPCM, you
> can't just play with the file, decode, re-encode and still have the same
> result unless you don't change anything anywhere in the file.

I don't know enough about ADPCM to comment, though, and I would be glad if someone more knowledgable than I am would offer some insight.