Louder Audio Advice?

[Ookamichi]

So I'm using HandbrakeCLI 0.10.1 and wish to make audio as loud as possible.

I know that the following command switches
--gain
Amplify or attenuate audio before encoding
--normalize-mix
Normalize audio mix levels to prevent clipping


Now my question is can I use both such as

--gain 6dB --normalize-mix

would it make a difference or would it cancel each other out?

[pandy]

Go for compandor (dynamic compression)

[Ookamichi]

Go for compandor (dynamic compression)

So this is the best over gain and normalize?

-D, --drc <float> Apply extra dynamic range compression to the audio, making soft sounds louder. Range is 1.0 to 4.0 (too loud), with 1.5 - 2.5 being a useful range.

Can you explain? thank you

I learned searching that it only works on when the source audio is AC3

my working files are mkv format with AAC audio
Audio
ID : 2
Format : AAC
Format/Info : Advanced Audio Codec
Format profile : LC
Codec ID : A_AAC
Duration : 23mn 42s
Channel(s) : 2 channels
Channel positions : Front: L R
Sampling rate : 48.0 KHz
Compression mode : Lossy
Delay relative to video : 9ms
Default : Yes
Forced : No

[AlexKane]

The vast majority of films contain a multichannel soundtrack that has been mixed and produced for theatrical installations. The dynamic range in these soundtracks is pretty enormous with the dialogues averaging at -27 to -31 db RMS, background sounds well below that, and the foreground sounds sitting above -31 db RMS. You effectively have a dynamic range of ~60 db with your dialogue sitting somewhere in between. This is why you can't comfortably listen to the soundtrack in quiet, low noise-floor environments. It is essentially incompatible with your environment.

The stereo encoded stream within your mkv, is a downmix from the original multichannel soundtrack present in your source. It contains the aforementioned ~60 dbs of dynamic range and is most likely scaled down in volume to prevent clips that downmixing will potential introduce.

What you need is something like the compander in sox. You can use it to compress the dynamic range above the dialogue (loudest peaks included) giving you enough room to normalize and increase the overall loudness.

[LoRd_MuldeR]

So I'm using HandbrakeCLI 0.10.1 and wish to make audio as loud as possible.

I know that the following command switches
--gain
Amplify or attenuate audio before encoding
--normalize-mix
Normalize audio mix levels to prevent clipping


Now my question is can I use both such as

--gain 6dB --normalize-mix

would it make a difference or would it cancel each other out?

I think the "--normalize-mix" option applies to mixing multiple channels only, e.g. when when doing Stereo-downmix of a Surround source. It means that the channel weights are normalized in way that ensure that the sum of all weights in the mix won't exceed 1.0. Thus "--normalize-mix" guarantees that there won't be any clipping (distortion) in the mix, even if all input channels have been at "full scale" already. If you are not mixing channels, this should have no effect at all.

At the same time "--gain" simply increases the volume of the entire file by a certain constant amount. Clearly, the maximum gain that you can add to the file without clipping (distortion) will be predefined by the "loudest" peak that exists in the file. Applying a compressor before adding gain will "shave off" the peaks, so you can add more gain without clipping - but it also ruins the dynamic range of the file (cf. "loudness war"). Try dynamic normalization (https://forum.doom9.org/showthread.php?t=170980) as a less "invasive" alternative.

[Ghitulescu]

Like everyone else here I believe you intend to increase the auditive perception of a commercial audio track.

Many (most? all?) audiotracks in Dolby have DRC value which you may use (in the menu of the player) to compensate for different environments.

see if this helps you


http://www.ac3filter.net/wiki/Using_dynamic_range_compression