Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Lame using higher bitrate for silence than for music (Read 5400 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Lame using higher bitrate for silence than for music

I encode my song to mp3 with V 0 setting and I notice that when in silence the bitrate of that frame become higher than when there is music. Is this bug? It goes all the way to 279 kbps in that silence and go lower when the music start again.
I will think about tomorrow's problem tomorrow

Lame using higher bitrate for silence than for music

Reply #1
That's strange.
Is it really silence? Low volume noise could be an explanation.
lame3995o -Q1.7 --lowpass 17

Lame using higher bitrate for silence than for music

Reply #2
That's strange.
Is it really silence? Low volume noise could be an explanation.

Yes it is silence. I can't upload the sample so I will just give you the link to the YouTube. The silence start at 2:38

https://www.youtube.com/watch?v=3hkZNgRUo8Q&app=desktop

OK I encode the song to mp3 again and mp3 bitrate is normal now. 224 kbps for silence frame
I will think about tomorrow's problem tomorrow

Lame using higher bitrate for silence than for music

Reply #3
I downloaded it and opened the audio in GoldWave.  GoldWave says The left channel is -49dB down and the right channel is -46dB down.    It's NOT "digital dead silence".

Lame using higher bitrate for silence than for music

Reply #4
I downloaded it and opened the audio in GoldWave.  GoldWave says The left channel is -49dB down and the right channel is -46dB down.    It's NOT "digital dead silence".

Well even though it not dead silence but mp3 still encode it higher than when it has sound. And I think that noise was noise floor and not actual music.
I will think about tomorrow's problem tomorrow

Lame using higher bitrate for silence than for music

Reply #5
Without getting in too far over my head in technical details, signals that are random are more difficult to compress than signals that are not random.  In the specfic case of mp3, Lame is expending a lot of data on the higher scalefactor bands because they are a lot more prominent in relation to the lower bands than they are in music. As an experiment you could try the -Y switch and/or a lower quality level to see if that makes any difference.

Slightly OT: I've been reluctant to upgrade Lame from 3.98.4 because I've been put under the impression that it will give me bigger files at my chosen quality setting and am operating from the why fix what isn't broken mentality. You could also see if this version treats the noise in this track differently.



Lame using higher bitrate for silence than for music

Reply #8
I downloaded it and opened the audio in GoldWave.  GoldWave says The left channel is -49dB down and the right channel is -46dB down.    It's NOT "digital dead silence".

Well even though it not dead silence but mp3 still encode it higher than when it has sound. And I think that noise was noise floor and not actual music.


If you're asking why encoding noise requires a lot of bitrate, it is because noise is relatively incompressible.

Lame using higher bitrate for silence than for music

Reply #9
OK thanks guys
I will think about tomorrow's problem tomorrow

 

Lame using higher bitrate for silence than for music

Reply #10
Quote
Well even though it not dead silence but mp3 still encode it higher than when it has sound. And I think that noise was noise floor and not actual music.
It's not noise, at least not ALL noise...    I amplified it and it's the music (or a tone) continuing to fade-out slowly.

In any case, there doesn't appear to be anything wrong with LAME.