[FFmpeg-user] WebM & Opus Question

[Vlad Negru]

 Hello!
Thank you very much for your answer.
What exactly do I have to do to compile opus myself?
Sorry, but I'm very new to using ffmpeg and don't really know a lot of things.
With Respect,
Vlad Negru     On Wednesday, June 12, 2024 at 09:57:24 PM GMT+3, Ferdi Scholten <ferdi at sttc-nlp.nl> wrote:  
 
 
On 12-06-2024 20:15, Vlad Negru via ffmpeg-user wrote:
> Hello!
> I have a question,
> I'm trying something pretty simple, I have a WebM file/container downloaded from YouTube containing only 16-Bit Opus audio. I want to copy and retain just the audio part since this container only has audio anyways, no video.
> The WebM file/container is called Input.webm and is located in the YouTube_Samples folder on my desktop.
> As shown in the attached picture, it is originally in a 16-Bit format which one would assume is using integer representation.
> For copying I'm just using these very simple commands:
> cd C:\Users\VLAD NEGRU\Desktop\YouTube_Samples
> ffmpeg -i Input.webm -c:a copy Output.opus
>
> However, the output file is now in a floating point format and my portable player won't play that.
> Just in case that it matters somehow for anything, I'm using Windows 10 Pro as operating system on my computer.
> Is it possible to change the default fltp (floating point) format when working with .webm and .opus files?
> Is it possible to have it set so that it retains the original 16-Bit integer format when just copying the audio from the container?
> If possible, then what is the command line argument for forcing the use of integers with .webm and .opus files?
You need to have a version of the opus encoder that is specifically 
compiled to use fixed point, the following is a quote from the opus git:

    "This implementation uses floating-point by default but can be compiled to
    use only fixed-point arithmetic by setting --enable-fixed-point (if using
    autoconf) or by defining the FIXED_POINT macro (if building manually).
    The fixed point implementation has somewhat lower audio quality and is
    slower on platforms with fast FPUs, it is normally only used in embedded
    environments."

So it is either fixed point, or floating point, but not both.
Most likely if you want fixed point you will have to compile opus yourself using the quoted settings.
_______________________________________________
ffmpeg-user mailing list
ffmpeg-user at ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-request at ffmpeg.org with subject "unsubscribe".