[CLAM] Questions about SMSTools (Frame Segmentation).

[Matthias Geier]

Thanks again.

> By looking at the code (and your explanation) I certainly think that is a
> bug due to unnecessary rounding operations. Your alternative for the
> termination condition looks much better to me at first sight and I can't
> think of any reason why the one in the code would be preferred.

Maybe it causes some unwanted side effects ... ?

> This termination condition is not very relevant in most cases because you
> don't have audio going until the last sample (you usually have zeros both
> at the beginning and end)

It may not be relevant for many applications, but in my case I am
splitting the original sound file into frames and I am using the
SMSTools data for further analysis of the audio data. Therefore I
obviously need the same number of frames for both audio and SMS data.

For now I'll add a single zero to the end of the soundfile as a
workaround, then my frame segmentation works as expected. And if in a
future release of SMSTools the behaviour is changed, I just have to
remove this workaround.

> In any case I guess the most sensible strategy is the first one but
> telling users
> to make sure there are sufficient zeros at the end of the sound file.

Also IMHO the first strategy makes more sense. On the other strategy
one would get negative centerTimes and centerTimes > Duration of the
sound.

Thank you very much for your replys,
Matthias