Recherche avancée

Sur d’autres sites (6247)

• ffmpeg - Replace one word audio in video at different specific timeline [duplicate]

  17 décembre 2019, par Siddharth

  This question is an exact duplicate of :

  • Can I extract an audio clip (a single word) from the video and insert at different points/timeline of same video using FFMPEG/libvlcsharp ?

  How can i replace one word (newyork to be replaced by california) throughtout the video where speaker says "Newyork". I want that "Newyork" word to be removed from the audio (while keeping video intact) and insert "california" throughout the video at different places.

  I have an audio file where speaker says "California" which i want to replace with "newyork" word.

  How can i do that using FFMPEG ?

• ffmpeg convert video to size 275x157 pixels [closed]

  21 janvier 2024, par degr

  I want to make video of exact size 275x157, but ffmpeg always make it 274x156.

  


  So far I try next list of commands :

  


  //it from my opinion, it should make vide of my size, and add padding if necessary, but result is same - 274x156
ffmpeg -i ScreenRecorderProject7.mkv  -strict -2 -filter:v "pad=275:157:(275-iw*min(275/iw\,157/ih))/2:(157-ih*min(275/iw\,157/ih))/2" WrathTrailer1334x750.mp4

//more advanced variant of previous command, not work also
ffmpeg -i ScreenRecorderProject7.mkv  -strict -2 -filter:v "pad=275:157:(275-iw*min(275/iw\,157/ih))/2:(157-ih*min(275/iw\,157/ih))/2" WrathTrailer1334x750.mp4

//it told width is not divisible by 2
ffmpeg -i ScreenRecorderProject5.mkv -strict -2 -vf scale=275x157 -aspect 1.75259 output.mp4

//in description it was mentioned it add 1px pad, but I feel it is wrong; result incorrect
ffmpeg -i ScreenRecorderProject5.mkv -strict -2 -vf pad="width=ceil(iw/2)*2:height=ceil(ih/2)*2" output.mp4


  


  Also I try to do so with movavi video editor, streamlabs and couple online video converters and always I'm getting wrong resolution. Cant understand what is wrong with that numbers

  


• Decode demuxed frames produced by ffmpeg wasm using webcodecs

  10 février 2024, par Przemek Gałęzki

  Im trying to use webcodecs VideoDecoder to decode frames demuxed by ffmpeg wasm, but it doesnt seem as straightforward like when I've used mp4boxjs which provides you with necessary data for each demuxed chunk to provide to decode(), or I just dont know how to use ffmpeg wasm to properly extract chunk data

  


  so my questions are :

  


  

  • how do I properly split whole demuxed file into chunks so that I can then provide those to EncodedVideoChunk ?

  


  • after I have those split, for each EncodedVideoChunk I also need to provide it with

    

    • timestamp, not sure if performance.now is all right or not ?

    


    • data, probably just buffer source of chunk — after i get those right

    


    • type, which is either key or delta, no idea when should be which

    


    • duration

    


    


  


  


  I also need to configure decoder, ive used ffprobe-wasm here, but it doesnt provide exact codec name and description.
Even if i know exacly whats the codec name and demux my mp4 video to annexb format which if im correct doesnt need description, im getting error : A key frame is required after configure or flush and that i need to fill out the description field anyways.
I will be happy to get some insights, it seems like a lot to dig into to get it working ..