Just a thought, from a knuckle-dragging biology scientist. TL;DR: I believe there is scope to make the hosting of a peertube instance even more lightweight in the future.

I read some time ago of people using #webAssembly to transcode video in a user's web-browser. https://blog.scottlogic.com/2020/11/23/ffmpeg-webassembly.html

Since then, I believe #WebGPU has done/is doing some clever things to improve the browser's access to the device's GPU.

I have not seen any #peertube capability that offloads video transcoding to the user in this way.

I imagine, though, that this would align well with peertube's agenda of lowering the bar to entry into web-video hosting, so I cannot help but think that this will come in time.

My own interest is seeing a #Piefed (activitypub) instance whose web-pages could #autotranslate posts into the user's own language using the user's own processing power... One day, maybe!

Thank you again for all your hard work; it is an inspiration.

#mastodon#webVideo #HLS #transcoding #video #decentralization#edgeComputing#webGL#W3C #activitypub#AI #ffmpeg #selfhosted #degoogle #mathstodon

Dear sound/audio folks and engineers,

I have a directory with 3.5GiB of audio files (chiefly opus & m4a) which are spoken word recordings.

Some of them are quite low, and some of them are quite dynamic such that it's a whisper at times and nearly a shout at other times.

I've processed a lot of them with #audacity's compressor filter or #ffmpeg (ffmpeg -i audio.m4a -filter:a "speechnorm=e=50:r=0.0001:l=1" audio-normalized.m4a), but there are some unprocessed files in the collection, which are a pain to individually find and fix.

Is there a way from the #CommandLine to detect the loudness and/or dynamic range of audio files so that I can automatically flag them for processing with ffmpeg?

Thanks!!

#audio #sound #SoundEngineering #AskFedi #HiveMind