Can mp3 waveform data be extracted?

Solar2D Daily Builds and Public Releases
#1

I want to play a mp3 file and display its waveform on screen.
I also want to draw some waveform controlled graphic on screen while the music is playing.
I wish there is API like this:
audio.getWaveForm(audioHandle, starttime, duration)

Is that possible?
[import]uid: 35642 topic_id: 7382 reply_id: 307382[/import]

#2

Bump.
No one interected in this? [import]uid: 35642 topic_id: 7382 reply_id: 26200[/import]

#3

I don’t think anyone is interected. Interested maybe, but definitely not interected. :wink:

We have received some requests on this feature. It is on the back burner. But there are other higher priorities.

C. [import]uid: 24 topic_id: 7382 reply_id: 26313[/import]

#4

Thanks carlos :slight_smile:
Sorry for the misspelling … [import]uid: 35642 topic_id: 7382 reply_id: 26877[/import]

#5

That is a shame, I am making an app with lots of really small sound files, would be good if i have a small pop-up with a soundcloud style waveform.

This isn’t a show stopper for the app, but would have looked cool.

Can anyone suggest which values to graph for this (as in which api calls) and I am happy to give it a try.

#6

If this doesn’t need to be “dynamic” (as in, playing some sound file and generating a graph on the fly, as it plays), I suppose you could gather this data from an audio utility of your choice, organize it within a Lua table per sound effect, and then use that data to output a visual graph in Corona (i.e., output a series of thin rectangle vectors representing the volume across the time span). Maybe this isn’t nearly what you had in mind, but just a basic suggestion. :slight_smile:

Brent

#7

That is definitely a good place to start.

In looking for ideas, I found this :- http://www.hisschemoller.com/2010/mp3-wave-display/

I will see what information i can get from the sound API to feed something like this.

Or make a utility to extract it before hand and load a database or something.

#8

That is a shame, I am making an app with lots of really small sound files, would be good if i have a small pop-up with a soundcloud style waveform.

This isn’t a show stopper for the app, but would have looked cool.

Can anyone suggest which values to graph for this (as in which api calls) and I am happy to give it a try.

#9

If this doesn’t need to be “dynamic” (as in, playing some sound file and generating a graph on the fly, as it plays), I suppose you could gather this data from an audio utility of your choice, organize it within a Lua table per sound effect, and then use that data to output a visual graph in Corona (i.e., output a series of thin rectangle vectors representing the volume across the time span). Maybe this isn’t nearly what you had in mind, but just a basic suggestion. :slight_smile:

Brent

#10

That is definitely a good place to start.

In looking for ideas, I found this :- http://www.hisschemoller.com/2010/mp3-wave-display/

I will see what information i can get from the sound API to feed something like this.

Or make a utility to extract it before hand and load a database or something.

#11

hi, did you get anything working to generate a waveform? I’m going to start in on this myself, curious how you are handling it.

thanks,

Jen

#12

Hey there,

We looked into the idea of using an offline utility, and merging that with the same techniques you use for voice sync, but in the end we did not pursue it.

Basically, our research indicated its not really something you can do with corona just yet, but the audio tools available are pretty extensive so we’re thinking it may come along in future.

We are talking with a client about doing it natively, which is possible currently, doubt that helps you though :slight_smile:

Jez

#13

hi, did you get anything working to generate a waveform? I’m going to start in on this myself, curious how you are handling it.

thanks,

Jen

#14

Hey there,

We looked into the idea of using an offline utility, and merging that with the same techniques you use for voice sync, but in the end we did not pursue it.

Basically, our research indicated its not really something you can do with corona just yet, but the audio tools available are pretty extensive so we’re thinking it may come along in future.

We are talking with a client about doing it natively, which is possible currently, doubt that helps you though :slight_smile:

Jez