Is it possible to split an mp3 file with fmod?
I want to make something like an subtitle workshop. But I don’t know if the seconds and milliseconds position in fmod matches if the time position in windows media player, or another program which I want to use to split the mp3 file. So before the problem that the position values differ about 30-40 ms. I think it is better to split it directly in fmod. But is this possible?
- maximilian.schaller asked 10 years ago
Has anyone figured that one out?
I’m trying to modify the realtimestitching example to use sections of an mp3. I don’t find any simple way of splitting a pre-loaded FMOD::Sound into a list of several FMOD::Sound, in order to play them in arbitrary order. Could someone help? I also looked at the sample-accurate sequencing from the wiki. It seems to be a more complicated version of realtimestitching, and still, doesn’t solve my problem of instantiating multiple Sounds from splitting a larger Sound. Thanks.
I’m still unclear as to what is the best way of accessing and playing chunks of a waveform with sample precision. Any suggestions anyone? Also the sample-accurate sequencing example does sound overlaps when I swap the example c_tiny.wav with drumloop.wav. Is this example working only with small files? Confused here. Thanks.
I don’t want to "split" the sound per se (like cutting audio into multiple chunks in memory), but I need to play small chunks of a loaded mp3 in arbitrary orders (say randomly). In raw C++, I’d do this by accessing various locations in the waveform, and by playing the given pre-defined chunk of audio, ultimately constructing a new, concatenated stream in real-time. In FMOD, it looks like this is done by double-buffering FMOD::Sound objects (or handles), and not by actually using pointers into a malloced waveform. Am I correct? In this case, I thought I’d need to instantiate these new FMOD::sound objects out of the original FMOD:sound (the entire mp3), even if there’s no actual copy of the audio. Should I do this differently using simple markers and no double-buffering? I don’t mind having all the audio decoded in memory for now. Any best example anywhere? Many thanks.
Please login first to submit.