I usually use the MeGui's HD Stream Extractor to extract the audio from my Blu-Rays and it works fine.
The problem I am having is when I check the log, its says its "removing dialogue normalization", and "could not fix remaining delay of ???ms" for .AC3.
If I am doing it to FLAC, it says "removing dialogue normalization", and "Reducing depth from 64bits to 24bits".
Why is it doing this and not leaving the audio as it is on the source file? Can I extract the audio from it exactly as it is on the Blu-Ray disc and keep the normalization?
+ Reply to Thread
Results 1 to 2 of 2
You probably don't want the dialogue normalisation. It's used to adjust the volume of the whole audio so the dialogue is at a certain level. It doesn't change the dialogue independently of everything else. It generally means a player will decrease the volume to some extent.
Lossless audio has a bitdepth. Lossy audio doesn't. When you convert lossy audio it has to be decoded (usually as 32 bit for lossy re-encoding) or decoded at a particular bitdepth for playback. There's possibly info in the source file specifying 64 bit for decoding if it's lossy, but eac3to converts to 24 bit for lossless re-encoding to prevent the file size becoming mental. When you encode with a lossy encoder there's none of that. The source is decoded and sent to the encoder, generally at 32 bit for lossy sources and maybe at the original bitdepth for lossless.
eac3to tries to replace delays and gaps in the audio with silence. For lossy encoding, the audio is encoded in frames, so it's probable the unfixed remaining delay is the duration of a single audio frame, which isn't anything to worry about, but if there's an unfixed delay in the log file you can use it when muxing the extracted audio if you want to.
If the source is AC3 and the output is AC3 and therefore not re-encoded, eac3to adds silent AC3 audio frames to the existing audio stream as required. When converting to another format there shouldn't be any limitation in that regard, so it can fix delays and gaps exactly. Occasionally I've worked with source files where there's been gaps or breaks in the audio stream, and as far as I know, all other extraction tools extract the audio without accounting for any gaps, causing the audio sync to change, although they usually write any initial delay to the name of the extracted audio so it can be used when muxing later on.
TSMuxer can extract stuff from m2ts files, or alternatively you can remux them with MKVToolNixGUI. MeGUI's File Indexer can extract the audio from MKVs. Remuxing might be a bit time-consuming for large source files though.
Last edited by hello_hello; 21st Dec 2019 at 09:06. Reason: Added some more info