Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[daveman]

This might sound like a dumb question but what exactly is the point of "analyzing audio"?

I looked in the wiki (does not provide any info) and in posts here and while people discuss issues with analysis of audio (and video), I am unsure why one needs to analyze it in the first place and once analyzed, what I can do with the information (and more specifically HOW do I do the things that one can do with this information)

Maybe some info about this would be helpful to use newbies?

Dave

[6233638]

Each track is analyzed to find the average volume, the peak level, and the dynamic range.
The average volume level and peak level are used when the Volume Leveling or Adaptive Volume (peak level) DSPs are enabled.
Dynamic range can be a useful indicator of track quality if you have multiple versions of the same album/track in your library. (generally more dynamic is better)

[daveman]

Thanks for your response.

However, looking at the information obtained from Analyze Audio, I see the following

Replay Gain
Peak Level
BPM (beats per minute I assume)
Intensity

Which of these refers to dynamic range?
Which field refers to Average Volume?

Thanks
Dave

[6233638]

Intensity was removed in MC19.

Analyze Audio should be updating:

BPM
Dynamic Range (DR) - A measurement of Crest Factor, which correspond to the measurements at the Dynamic Range Database
Dynamic Range (R128) - A measurement of the track's actual dynamic range; the difference between how loud the loudest and quietest parts are.
Peak Level (Sample) - The maximum level of any samples in a track. This should not exceed 0dB with lossless files.
Peak Level (R128) - The track is upsampled 4x and then the peak level is measured. This is known as the True Peak Level, and can exceed 0dB.
Volume Level (R128) - Volume level analysis conforming to the R128 spec.
Volume Level (ReplayGain) - ReplayGain 2.0 values derived from the R128 data.

[DoubtingThomas]

Also, the analyzed values are stored in the MC database and in the tags of each track.  None of the actual audio information is changed.

[javidan]

This might sound like a dumb question but what exactly is the point of "analyzing audio"?

Dave

Hi Dave,

I use audio analysis for the purpose of allowing MC to play my audio/video files at a controlled volume, so that when I set my physical volume to my comfort level, I no longer need to make changes to it. MC has already gone through all the audio and intelligently adjusts the volume to a suitable level just about as "loud".

When you analyse audio, MC is really seeking out what is low in volume and what is high in volume, which it can later use to control the volume.
This is known as volume leveling. The common old method used by most players is called "Replay Gain", but based on my own experiences it didn't really work as well as described.

The new improved method called R128 does seem to achieve it as I do find that once I set my music to play, I do not see the urge to turn the volume up or down any longer.

The technical details are obviously going to be beyond me to explain and several of the answers appear to be how it works.

I am just telling you why I use it...I am totally blown away because I use it for videos as well now. (And funny that, I used to ask if it was possible to analyse audio for videos, JRiver explained it wasn't needed so they do not do it. LOL)

There is more to it than I am trying to explain since there is the adaptive volume bit vs volume leveling and I hope someone would be able to help both you and me understand it without using technical terms.

[astromo]

This might sound like a dumb question but what exactly is the point of "analyzing audio"?

I looked in the wiki (does not provide any info) and in posts here and while people discuss issues with analysis of audio (and video), I am unsure why one needs to analyze it in the first place and once analyzed, what I can do with the information (and more specifically HOW do I do the things that one can do with this information)

Maybe some info about this would be helpful to use newbies?

Dave

Have you checked this sticky in the Win MC19 forum?
New Features in MC19

In it there are these 3 key entries (daveman, you probably have spotted these entries but for others who follow, some links should serve to focus the Q&A):
NEW: Improved audio analysis and volume leveling (R128)
NEW: Adaptive Volume
NEW: Analyze audio for videos

The above covers quite a lot but just to demonstrate the wonders of Interact, here are some personal faves to provide some further background:
http://yabb.jriver.com/interact/index.php?topic=80199.0
http://yabb.jriver.com/interact/index.php?topic=77129.0
http://yabb.jriver.com/interact/index.php?topic=20920.0

There's probably more out there. It's just a bit of a journey ..

[mwillems]

Dynamic Range (R128) - A measurement of the track's actual dynamic range; the difference between how loud the loudest and quietest parts are.

A little additional info about Dynamic Range (R128): the spec compares the difference between the track's 95th percentile for loudness (to prevent a single very loud sound from setting the peak) and the 10th percentile for loudness (to screen out silences or fade outs), so it isn't literally the difference between the loudest and quietest parts of the track.  Which is to say if you deducted the Dynamic Range (R128) from the Peak Level (R128), you wouldn't necessarily get the volume level of the quietest part of the track or even very close to it (if the peak was set by an unusually loud drum hit, for example).  

I previously wondered how they dealt with the issue of track silences, and found that portion of the spec to be interesting:
http://tech.ebu.ch/docs/tech/tech3342.pdf