How to get sample depth from Sound?

[Владимир Коваль]

Hi!

I do:

snd = parselmouth.Sound(wav_path)

How can I get sample depth from snd (8 bit, 16, 24, etc. or 1-2-3 bytes)?

For now I use:

with wave.open(wav_path, 'rb') as file:

    sample_width = file.getsampwidth()

but if it possible only with parselmouth library - it will be great!

[yannick...@gmail.com]

Hello!

Parselmouth abstracts this away, when reading (just like Praat itself, as Parselmouth mainly provides the Python interface). All audio files get read and immediately converted to 64-bit floating point numbers between -1 and 1.

If you need to know this information about the actual raw audio file, I think that your approach of using the built-in wave library seems correct. However, if the rest of your acoustic/phonetic/... analysis is done with Parselmouth, this info shouldn't matter too much?

Kind regards

Yannick

[Владимир Коваль]

Got you, thanks!

Indeed, this is not a problem, just some perfectionism stuff :)

суббота, 15 января 2022 г. в 19:01:54 UTC+3, yannick...@gmail.com:

[yannick...@gmail.com]

Ah, yes, I perfectly understand :-)

So, yes, I would guess Praat doesn't include such information because it wants to provide functionality on a higher level.

But then this is exactly why I wanted to develop Parselmouth, as it allows to combine with other Python libraries ;-)