Feature Request: Smart Volume Leveling for Streaming Services

Hello,

I am an owner of the Eversolo DMP-A6 Master Edition Gen 2 and would like to suggest a feature for future firmware updates.

When using streaming services such as Deezer, TIDAL, Qobuz and others, the loudness of tracks can vary significantly. Even with existing loudness standards, different recordings, albums, and remasters are often played back at noticeably different volume levels.

I would like to suggest an optional feature called Smart Volume Leveling.

The idea is very simple:

The device should automatically and smoothly adjust only the overall playback level between tracks while preserving the original sound quality.

Important requirements:

• No equalization (EQ).
• No changes to frequency response.
• No dynamic range compression.
• No alteration of the recording’s character.
• No degradation of audio quality.

The feature should work like an intelligent volume knob:

• Quiet tracks are slightly amplified.
• Loud tracks are slightly attenuated.
• Adjustments are smooth and transparent.
• A comfortable and consistent listening level is maintained during long listening sessions.

This would be especially useful for users who listen to large playlists, mixed genres, internet radio, and streaming services.

It is important that this feature remains completely optional and can be disabled by users who prefer bit-perfect playback.

I am not suggesting compression or dynamic processing of music. The goal is simply automatic gain adjustment between tracks to provide a more consistent listening experience.

Many Eversolo owners use their devices not only for critical audiophile listening but also for everyday music enjoyment. Such a feature could significantly improve user experience without compromising sound quality.

Thank you for your attention and for the continuous development of the Eversolo platform.

 

there is already a (more or less) standardised way of doing it, and it is lies in the loudness tags (aka ReplayGain).
There are tools to do that for your local library, but in the case of streaming services there is (of course) no way of doing it; unless the tracks themselves present such information upon stream-start...

There is one thing, in all of your description that does not fall into place, IMHO: how could you possibly know whether this track is too loud, or the following one is going to be too quiet, when it comes to streamed tracks?!?

Moreover... in order to determine the overall loudness of a track, the entirety of it needs to be scanned... and for a stream this is impossible...

Genuinely asking, as it would be an interesting topic to explore...

 

Sounds like old school audio, we probably had it right the first time, A good number of those requested features when you think about it, we already had to a standing degree, depending on what you’re trying to accomplish naturally. While in view of your post, sorry I to had lol Remember the old straight line preamps tube or solid State no difference no EQ’s what went in is what came out untouched on altered , yes absolutely most times you would always retain the use of gain and level controls, balance, speaker placement simplicity works and worked well I have a tendency to forget we’re in the New World though the song remains the same. However, your post did bring back old listing sessions long before you need an update an upgrade with the exception of a wore out tubes and a few capacitors , along with a top notch controlled audio source very minor issues in overall performance adjustment within your listening domain, I absolutely believe most of what you are asking or Seeking is possible, but not within this mode of travel, not quite ready in personal opinion only , but hey some folks love it , theirs really not much to compare it to today , well not at these cost ratios at present , I remember when a good MC cartridge cost more than that , one of the best systems sources I have ever heard to date , half inch R to R at 30.IPS . I absolutely agree and do retain your point of view ,like a good buddy stated, today I get the feeling , we’re playing with toys, today at times , however excellent post I absolutely enjoyed the read,;-) you appear to have had some excellent experiences within the recorded of reproduced true and accurate untouched Audio Arts . Theirs only one original often copied, but never duplicated,

please have an excellent weekend ……

 

Hmmmm
leveling loudness seems like a crazy idea. Leveling gains based on a codec or streaming service seems more appropriate.
I don't want my songs sound just as loud. Soft songs should stay soft and hard songs should sound loud. That is the ff-ed up reality of playlists. you can not combine the soft with the loud. Leveling them out is a flip of the finger to the musician.
Some streamers do normalize their tracks. which means they level out the max outputs. So it won't matter if a track is based on a CD or something else.
The eversolo could/should do something similar The max outputs of streamers and local inputs/tracks should be able to be leveled.
For local files the replay gain is a standard tag to do so. It would be nice if tidal came with a gain and spotify and deezer etc. every steamer or input should have a gain setting.
I don't want eversolo to change the behaviour of tidal/spotfy/qobuz just apply a gain adjustment. (0db as default setting )

If tidal/qobuz is kind enough to enlighten the api with 'replay gain' info. The eversolo could/should apply this.

 

Thanks for your kind words and for sharing your experience. I enjoyed reading your perspective. Have a great weekend too!

 

some folks view leveling with gain structure or input level, although somewhat closely related personally, there should be both input gain structuring and level output gain attenuation, in my miss spent youth I was actually shown and taught not to confuse gain input with gain output, with the mater volume output, I believe Manley, AUDIO Labs still manufacturers and assortment of the units, depending upon needs and request, but naturally mostly unheard of for the average Audio hobbyist. These have personally only been my observations, although still fashionable somewhat old school today some of the source material compression is almost unbelievable do you remember the loudness wars Somewhat of an example, I would imagine in view of days gone by naturally…it appears to have been somewhat better in audio output in recent years,

 

@svideo it is indeed a (if not THE) problem of playlists... more emphasised if you listen to serious stuff of the late 80s, and some of the - late? - 90s (aka the loudness war...)
The ReplayGain tags are indeed helpful for the local library; they can provide also loudness leveling based on the album, so that a quiet track will ALWAYS be quiet; even when played in a playlist/compilation.
This said, never used the functionality, because I do not listen to playlists. And if I ever will, I'll just scan my entire library, generate the tags, and let my player use them.

The streaming guys will have to wait, or rely to the service for leveling the playlist before plyback.

 

if one takes note or interest within the ability and using DBX. There philosophy, I believe in the following order louder sounds are expanded to become somewhat louder. Softer background sounds are still expanded, but in such a way to coincide within the loud passages, yes believe it or not. These older units are still around and there are a few guys that still use them on a daily basis they can work quite well, but takes a lot of patience, knowledge and understanding and overall operation Although I believe Nutul is quite correct it’s an entirely different playing field as of today, although may be gone, but certainly not forgotten, Nutul within all likelihood, I believe certainly understands let’s say the inner workings within certain devices that may be deemed obsolete by some and still cherished by others also these Particular devices can still be quite useful I believe most likely inserted after the output within your Eversolo unit, then processed and passed along to existing units or devices, however, there can be great deviation and variation within use of, but to a certain extent absolutely will work, now you are starting to get into it noise reduction, and everything that coincides along with it, yes in certainly can be a productive activity in practical use, but extremely difficult to keep all spectrum frequencies in equal operation from purely a sonic standpoint within final outcome…

 

I would be satisfied already if the "replay gain" function would work with TIDAL.
TIDAL provides the information (I can tell because it is shown in the display for a few seconds for each song), but the DMP-A6 does not seem to use it.
I tried replay gain settings "off", "track gain" and "album gain", nothing ever changes, the (sometimes great) volume variances remain.

 

Many modern streaming services (Spotify, Tidal, etc.) already have their own volume normalization.

 

Again Speaking on audio leveling output, let’s take a streaming service or also using other inputs / outputs from many other devices or sources, that may use our employe volume normalization sounds very useful in standardized level output within using any source, I have often found some particular music may seem like it’s sonically outputs gain at a lower or higher level of execution may in fact be due to certain artist, albums songs, some studios intentionally recorded and produced within specific gain output structures some songs that sound much louder, in gain output than others that were produced and recorded I would believe for various reasons perhaps do to the type of music itself not to play or sound in gain output just as loud and outputs from another opposing artist, album song, etc. in personal opinion, with auto leveling availability, some specific music was recorded with the intention of an overall sound or sonic presentation in comparison to perhaps another that was produced and recorded at a much higher level within recorded gain structure input that would naturally affect output in either case, although gain leveling structure may be possible the music or song that was recorded at a lower input naturally, we’ll have a softer Sonic output, regardless of leveling structure as that was the intended reason in doing so,

when listening to a softer sound structure within the recording, no matter how much level may be adjusted or deviated either way I would certainly believe there will still be an absolute difference within total output of volume or gain in regarding within specific recordings.if there is a very soft sounding sonic recording, increasing the level up on a recording that was produced and recorded with a much higher gain structure in output or lower. Would most likely still retain somewhat of a higher or level output within overall gain due to the nature within the way, the actual song or Recording were originally intended to be heard and produced and recorded , In my personal experience and findings. however I most certainly see your concern in understanding absolute equal level control in theory, but not in reality just my personal opinion, please have a wonderful weekend ……

 

It's available to enable Replay Gain in the Playback settings, once the service support.

 

Yes, thank you, I know. I have activated it. And TIDAL sends replay gain information, it is even shown on the display for each new track starting to play.

My point is, the DMP-A6 ME Gen. 1 does not seem to use this replay gain information provided by TIDAL. The variances ate not compensated.

Example: If I have 2 TIDAL Tracks with enormous volume variance (almost 10 db).
Regardless if I switch Replay Gain function on or off in the DMP-A6 settings, the volume variance between those tracks as I can hear it remains unchanged. It doesn't seem to work. A bug maybe?

EDIT:
This just came to my mind: I have the DMP-A6's volume controll disabled, it's on passtrough/100% permanently because the amplifier controls the volume.
Maybe for this reason the track gain function does not work for me?
I need to check this later and will report back here about the results.

 

No, the replaygain function is available under the passthrough mode.

Please record a video of the playback screen after enable the replaygain source mode, I'll forward to our team for a troubleshooting : )

 

Thank you, I will do so.

 

Hello again, Kim, I tried to upload the file (~8MB) here, as a .mp4 or a .mov File, but after upload I always get the message that the file extension is not allowed...

ah, never mind, I could upload it after renaming it to .txt.
Please rename back to .mp4 after download. Hope this will work.
Also hope that this is what you wanted to see.

Thank you
Thomas

 

Thank you, Thomas.
Yes, it works, I have received the video.
However, we didn't replicate this issue. Could you please confirm the firmware version on your A6?

 

Thank you, Kim

I am currently on business trip and cannot check, but it should be latest firmware for DMP-A6 Master Edition Gen 1, i have set it to auto-update. Beta versions excluded.

I will check and post here when I am back home.

 

Thanks : )
If you can record a video of the issue it'll help us to troubleshoot.

 

When some day in the distant future - probably in 10 years - Eversolo will support Atmos, then there should be a way to get equal levels from stereo and Atmos tracks. Right now, when listening to Amazon Music, Atmos tracks are so much quieter than stereo (I guess about 18db) that it´s impossible to listen to albums, which have stereo and Atmos tracks. And it´s nearly impossible to listen to Atmos really loud, anyway. Even with a Cinema 30. No problem btw, when using Atmos from BluRay.