We analyzed the loudness of YouTube videos in Japan

Japanese YouTube Loudness Histogram

We analyzed the sound pressure of the YouTube video in Japan .

On YouTube, loudness normalization is introduced. Loudness normalization on YouTube lowers the volume of the video whose sound pressure is too high, but does not raise the volume of the movie whose sound pressure is too low.

If you have lots of videos with too little sound pressure on YouTube, you can increase the volume played on YouTube by raising the sound pressure to the extent that sound quality is not impaired.


How much movie is too small sound pressure on YouTube?

As mentioned above.

Does the sound pressure drop if the YouTube video length is long?

As the length increases, the probability that the peak of the waveform will increase by chance will increase. If you use a limiter, you can suppress those peaks, but if you do not use a limiter, the sound pressure may decrease as the length increases.

From this, if you examine the relationship between length and loudness, it seems to be a material to judge whether the limiter is used or not. There are other factors related to length and loudness, so we can not make a judgment alone, but once we examine the relationship between length and sound pressure.

Analysis target video

A. Japan's Top YouTuber

Almost all videos of "Hajime" channel

B. Japanese music channels

Almost all videos of "Lantis Channel" channel

C. Japanese TV station channel

Almost all videos of "AbemaTV Official YouTube" channel

* Videos before 2018/12/8

※ Because almost all of the movies have failed analysis

※ The video list is described in Appendix

Analysis measures


It is an indicator of the sound pressure. Calculated with ITU-R BS.1770-3.

Loudness Range

It is an indicator of dynamic range. We calculated and calculated the window length and overlap length of EBU TECH 3342. Window length 0.4 s, overlap length 0.3 s.

Analysis Result


Loudness histogram

Japanese YouTube Loudness Histogram

Loudness cumulative density distribution

Japanese YouTube Loudness CDF

Loudness time series

Japanese YouTube Loudness Time series

Loudness Range

Loudness range histogram

Japanese YouTube Loudness Range Histogram

Loudness range cumulative density distribution

Japanese YouTube Loudness Range CDF

Loudness range time series

Japanese YouTube Loudness Range Time Series

Relationship between loudness and loudness range

Loudness vs Loudness range scatter plot

YouTube Loudness vs Loudness Range

Loudness vs. Loudness average and standard deviation

Japanese YouTube Loudness vs Loudness Range error bar


Length Histogram

Japanese YouTube Length Histogram

Length cumulative density distribution

Japanese YouTube Length CDF

Length time series

Japanese YouTube Length Time series

Relationship between length and loudness

Length vs Loudness Scatter plot

Japanese YouTube Length vs Loudness

Length vs Loudness average and standard deviation

Japanese YouTube Length vs Loudness error bar


How much movie is too small sound pressure on YouTube?

Looking at "Loudness cumulative density distribution", except for "Lantis", the loudness of 90% or more of the video is -14 dB or less.

Since the loudness calculation formula used this time is different from that of YouTube, I can not show the line that loudness normalization is applied any more, but as I think that -14 dB is sufficiently small as reference here , since sound pressure other than Lantis If you raise it, there seems to be a possibility that the volume at the time of playing on YouTube will rise.

Does the sound pressure drop if the YouTube video length is long?

According to "Loudness vs Length Average and Standard Deviation", such a fact is unlikely.


YouTube video analysis result in Japan (tsv)


I analyzed the sound pressure of the YouTube video in Japan.

"Haruan" is a Japanese YouTuber who is fully compliant with the loudness normalization!

" Haruan " may be YouTuber which is fully compliant with loudness normalization .

Loudness normalization fully compliant What is YouTuber?

YouTuber understands YouTube 's loudness normalization specification and uses it well .

What is Loudness Normalization?

YouTube automatically adjusts the volume between videos.

Have you seen "detailed statistics" from the right click menu of YouTube videos?

youtube video stats

Please pay attention to "Content loudness" in this. This value is a reference when YouTube performs loudness normalization.

If "Content loudness" is positive, YouTube will lower the volume. If it is minus, the volume remains unchanged.

Two facts about sound

To make good use of loudness normalization, the following two facts are important.

A. There is a trade-off relationship between sound pressure and sound quality

B. Larger volume sounds better

How to make good use of loudness normalization

On YouTube, since loudness normalization works, there are points where the sound volume does not rise even if the sound pressure is raised . It is the point where the above "Content loudness" becomes 0.

Since the sound pressure and sound quality are in a trade-off relationship, increasing the sound pressure so that the "Content loudness" becomes 0 or more on YouTube, the volume does not rise, so the larger the volume, the better the sound is heard Instead, only the sound quality goes down .

So, the best solution when raising a video on YouTube is to set "Content loudness" near 0 or below 0 .

"Harukanaru" video content loudness "

Please take a look at "Content loudness" of recent Hara video. I think that every video is near 0 dB .

Video of 2018/12/02 ("Content loudness" 0.0 dB)

Video of September 23, 2018 ("Content loudness" 0.0 dB)

In other words, Haruan's movie is the best balance of sound pressure and sound quality on YouTube. It is unusual for it to happen unintentionally, so it may be intentionally done.

However, in the video before a while, "Content loudness" deviates from 0 dB .

2018/05/25 movie ("Content loudness" -1.7 dB)

Video of 2017/11/12 ("Content loudness" - 8.5 dB)

Recently, it may correspond to the loudness normalization.


It was an article that YouTuber " Haruan " might be fully compliant with the loudness normalization .

For YouTuber applications and news applications, YouTube's loudness normalization criteria are sufficiently low, so it may be natural to respond like SEO from now on .

YouTube's loudness normalization algorithm

Estimated Weighting Curve Used for YouTube Loudness Normalization

In order to raise the highest sound quality video to YouTube, you need to know the YouTube loudness normalization specification.

However, YouTube's loudness normalization specification is not published. Some people have already been investigated, but specific calculation formulas are not known.

I tried to estimate the formula for loudness normalization on YouTube.

YouTube loudness normalization specification

The following is a summary of the survey results.

The loudness normalization is performed in a manner that the loudness of the sound source is adjusted to the loudness target value as much as possible within a range where the peak does not clip.

The loudness of the sound source is calculated with its own specifications, but by replacing the weighting curve of Short-term loudness of EBU TECH 3341 with the following and taking the maximum value of Short-term loudness, it is possible to obtain the accuracy of 1 dB Can be approximated.

Estimated Weighting Curve Used for YouTube Loudness Normalization

Research policy

We will investigate the framework of YouTube's loudness normalization in detail and the details of loudness calculation.

A large frame of YouTube's loudness normalization

I think that probably it looks like the following when referring to here .

The loudness normalization on YouTube is done in a way that the loudness of the sound source is adjusted to the loudness target value as much as possible within the range where the peak does not clip. When written with an expression, it becomes the following.

Compensation (dB) = Min (- Peak, Target - Loudness)

Peak is the peak of the sound source, Loudness is the loudness of the sound source, Target is a constant, the loudness target value, and Compensation is the correction gain. The overall volume changes uniformly by the amount of Compensation.

Right click on a YouTube video and the content loudness seen from the detailed statistical information is equivalent to Loudness - Target.

Loudness calculation formula on YouTube

YouTube's loudness calculation formula seems to be using its own one. So, I need to guess.

Consider the following model with reference to ITU-R BS.1770-3.

Equalizer -> Cut by window -> Convert to LUFS -> Gating -> Aggregation


Weight each frequency by an equalizer.

In previous experiments, K-weighting adopted in ITU-R BS.1770-3 and other popular weighting did not apply, so estimate direct frequency characteristics.

Cut by window

Cut out the waveform with the Rect window.

Window length and overlap ratio are parameters.

For reference, the momentary and integrated parameters of ITU-R BS.1770-3 and EBU TECH 3341 have a window length of 400 ms and an overlap length of 100 ms (the overlap ratio is 75%). The short-term loudness parameter of EBU TECH 3341 has a window length of 3 seconds and an overlap length of 2.9 seconds or more (overlap ratio is 96.7% or more).

Convert to LUFS

Calculate the RMS of the extracted waveform and convert it to LUFS with Log 10 (RMS).

It also corrects to be 0 with stereo 1000 Hz sine wave. The correction amount for ITU-R BS.1770-3 is -0.691 dB.


In order to eliminate the influence of silence time on loudness, we discard small sounds among multiple RMS values ​​obtained by cutting out.

Refer to ITU-R BS.1770-3 and EBU TECH 3342 and perform Absolute threshold gating and Relative threshold gating.

The parameters are the respective Threshold values. I also try patterns that do not do gating.

For reference, the parameters of ITU-R BS.1770-3 and EBU TECH 3341 are Absolute Threshold -70 LKFS and Relative Threshold -10 dB. Parameters for calculating the Loudness Range of EBU TECH 3342 are Absolute Threshold -70 LKFS and Relative Threshold -20 dB.


Take the average or maximum of multiple RMS values ​​remaining in Gating.

ITU-R BS.1770-3 takes an average, but it seems there is a possibility of using the maximum value of Short-term according to this .

Test video used for parameter estimation

Prepare a test movie to estimate the parameters of the loudness calculation model.

According to here , it seems that there is a possibility that loudness normalization will not be applied if there is not a certain number of playback numbers, or it will not be applied unless some time has elapsed since posting. Without preparing test videos on their own, there are enough playback numbers, select some of the existing videos that have been posted enough times, and make them test videos.

A list of test videos is described in the Appendix.

Equalizer parameter estimation

By using a sinusoidal test movie with a constant volume, you can eliminate effects other than equalization on loudness. Using this we first estimate the frequency response of the equalizer.

For the sine wave sound source of various frequencies, measure the content loudness on YouTube and estimate the frequency characteristics by taking the difference from the RMS of the sound source. The estimation result is below. For detailed data please see Appendix.

Estimated Weighting Curve Used for YouTube Loudness Normalization

The result was unstable, for example, the results were different depending on the animation even at the same frequency above 16 kHz, so in the following discussion, we will only use data below 15 kHz. Extrapolate with linear interpolation for 44 Hz or less and 15 kHz or more.

Parameter estimation other than equalizer

Next, fix the frequency characteristics of the equalizer and estimate parameters other than the equalizer.

Calculate the loudness of various videos with various parameters. Compare with the loudness (Content Loudess) calculated by YouTube and look for the parameter with the least error. The test video list is described in the Appendix.

Parameter list

Window length400 ms, 3 sec
Overlap ratio75%, 96.7%
Absolute thresholdNone, -70 LKFS
Relative thresholdNone, -10 dB, -20 dB
Aggregationmean, max

Result list

ParametersEstimated Target (LUFS)Error Stddev (dB)Error Max (dB)
abs threshold none, rel threshold none, window 0.4 sec, overlap 75%, mean-16.154494085.5125536210.73290254
abs threshold none, rel threshold none, window 3 sec, overlap 96.7%, mean-14.976814844.90827864611.91484089
abs threshold none, rel threshold - 10 dB, window 0.4 sec, overlap 75%, mean-13.949879233.9543709897.389401665
abs threshold none, rel threshold - 10 dB, window 3 sec, overlap 96.7%, mean-13.686847213.6840072747.647167492
abs threshold none, rel threshold - 20 dB, window 0.4 sec, overlap 75%, mean-14.498314374.5312554069.145055115
abs threshold none, rel threshold - 20 dB, window 3 sec, overlap 96.7%, mean-14.016606914.0487230579.667181199
abs threshold - 70 LUFS, rel threshold none, window 0.4 sec, overlap 75%, mean-16.154494085.5125536210.73290254
abs threshold - 70 LUFS, rel threshold none, window 3 sec, overlap 96.7%, mean-14.976814844.90827864611.91484089
abs threshold - 70 LUFS, rel threshold - 10 dB, window 0.4 sec, overlap 75%, mean-13.892175143.9115433187.447105751
abs threshold - 70 LUFS, rel threshold - 10 dB, window 3 sec, overlap 96.7%, mean-13.665658633.6660259727.668356069
abs threshold - 70 LUFS, rel threshold - 20 dB, window 0.4 sec, overlap 75%, mean-14.471706544.523919589.171662946
abs threshold - 70 LUFS, rel threshold - 20 dB, window 3 sec, overlap 96.7%, mean-14.005124264.0383895339.678663846
abs threshold none, rel threshold none, window 0.4 sec, overlap 75%, max-8.9937215021.1069610212.968119771
abs threshold none, rel threshold none, window 3 sec, overlap 96.7%, max-10.312464140.901435591.746039964
ITU-R BS.1770-3-10.3931764511.0314121233.14216451

Parameter combination with the least error was window size 3 seconds, overlap rate 96.7%, Max aggregation, standard error of error was 0.9 dB, maximum error was 1.7 dB. It is the maximum value of Short-term loudness of EBU TECH 3341. The loudness target value is -10.3 LUFS.

With this, you can estimate the loudness calculation method of YouTube.


Equalizer parameter measurement result (tsv)

Parameter estimation data (tsv) other than the equalizer


ITU-R BS.1770-3



I tried to verify Youtube's loudness normalization.

change history

2018/12/09 Fixed a calculation error (latest version)

2018/12/7 first edition


I looked up the formula for loudness normalization on YouTube. I found an expression that can be approximated with an accuracy of about 1 dB.

How to use Equalizer APO

We will show you how to use Equalizer APO.

Equalizer APO is Windows software that can apply audio processing to all sounds played on PC . For audio processing, you can use volume, equalizer, convolution of IR (impulse response), VST and so on.

Install Equalizer APO

If you are using 64bit Windows, download the 64bit version here and install it.

※ If you are using 32bit Windows, please download the 32bit version from here .

Equalizer APO setting screen

Open Equalizer APO 1.2 / Configuration Edtor from the Windows start menu.

equalizer apo

Equalizer with Equalizer APO

Equalizer can be applied in the arrow part of the screen below.

equalizer apo equalizer

Change the volume with Equalizer APO

You can adjust the gain with the arrows on the screen below. Even if you just cut it with an equalizer, the waveform will change and the peak will pop out so it may be better to lower the gain to prevent clipping.

equalizer apo gain

Use VST with Equalizer APO

To prevent clipping, try inserting the limiter at the end of the Equalizer APO.

Click the plus icon. Please choose Plugins / VST plugin.

equalizer apo add vst

The VST container was added at the very end of the process.

equalizer apo vst

Click the file icon.

equalizer apo open vst

Since the VST selection screen appears, select the VST file (extension .dll). Only VST 2 is supported. Since there is no bridge function, only VST with the number of bits matching Equalizer APO can be used.

equalizer apo choose vst

We were able to load VST.

equalizer apo vst loaded

Please click "Open panel" button.

equalizer apo vst open ui

The UI is displayed. Unlike DAW, please note that the setting will not be reflected unless you press "OK" button or "Apply" button. If "Apply automatically" is checked, settings will be automatically reflected as with DAW.

equalizer apo limiter

Possibility of Equalizer APO

Equalizer APO can be used in various ways.

Because the sound of YouTube movie is so loud that it cuts out high, or because there are many variations in the sound of YouTube movies so normalize it, because the dynamic range of the movie is too big to listen at home so compressing, speaker frequency characteristic correction And distance correction and so on.

Equalizer APO has made it possible to receive the benefits of VST not only for music producers but also gamers and YouTube viewers .


We introduced how to use Equalizer APO.

Actually, Equalizer APO taught foreign users that we use ITDPanner VST of our company for speaker correction. Please let me know if you want this VST!

Edit videos on iPhone

Here's how to edit videos on iPhone. This time I will do it with free iMovie.

Install iPhone video editing application iMovie

Install iMovie from the App Store.


Create a project

When you start it, it becomes the screen on the left, so press the plus sign on the project tab. Then select a movie.

iMovie Project    iMovie Add Project

Load movie

Select the material video. Multiple selections are possible.

imovie add video before    imovie add video

Cut video

When you tap a movie, an editing menu like the left picture appears. After choosing scissors and choosing Split, you can cut the video at the point you select like the right image.

     iMovie Edit Cut After

Sort videos

Touch and hold the movie to change the order of the movies by drag & drop.

iMovie Edit Move

Adjust the volume of movies

You can adjust the volume by tapping the video with a choice and selecting the speaker in the Edit menu.

iMovie Edit

Export movie

By pressing the completion button at the upper left of the screen and pressing the send button at the bottom of the screen, you can export the movie to various applications. For this time, save to the photo library, so choose Save Video. When you select export settings, the export will start.

iMovie Export  iMovie Send  iMovie Export Settings  iMovie Exporting  iMovie Exported

Increase the quality of audio (publicity)

Using our AI Mastering, you can increase the quality of the edited movie. Please try it with a video made with iMovie!

For details, please click here .


I showed you how to edit videos on iPhone.

The efficiency of video editing and the quality of the resulting video will vary greatly depending on which application you use. Therefore, selecting an application is important.

I will introduce a video editing application other than iMovie at a later date.

FAQ - Frequently Asked Questions

I summarized the frequently asked questions of AI Mastering .

Please teach me how to use it

Please see here for the web service version.

Please click here for the smartphone version.

Please tell me the price

Please see AI Mastering Pricing .

Can I use it for free?

The web service version can be used free although it is limited. For details, please see AI Mastering Pricing .

The smartphone version can be used free of charge.

Can I use it from a smartphone?

Although there are fewer features than the web service version, there is a smartphone version specialized for mastering movies. Please download from the App Store link below. Please see here for usage.

Can it be used by a corporation?

It is available.

Please register from AI Mastering . ※ However registration for the number of people required is required

Can you incorporate the automatic mastering function into your own service?

Is possible.

Cooperation by API and execution binary is possible. Please inquire details.

Can I do noise reduction?

It is possible with the smartphone version.

The smartphone version can be downloaded from the following App Store link. Please see here for usage.

Can I mix karaoke sound sources and vocals?

The current situation does not correspond, but it is scheduled to be implemented.

Is pitch correction possible?

Current situation does not correspond. It is under consideration.

Can I master my videos?

Both web service version and smartphone version are possible.

Please upload the video file to AI Mastering. Separate the audio and video parts of the movie, master only the audio part, and combine.

When downloading a movie, please download it from the download button under the movie displayed on the mastering result screen instead of the normal download button.

Is the uploaded sound source public?

It will not be published.

Does the uploaded sound source disappear?

It automatically disappears from the old one in about 3 days.

Can I use it without having a credit card?

If you can purchase V Preca at a convenience store, you can subscribe to the premium plan without a credit card. Purchasing for each mastering is also possible.

Can I sign up for a premium plan with PayPal?

can not.

Implementation is under consideration.

I would like to know the mechanism and internal algorithm of AI Mastering.

It is only an overview, but it is summarized here . Please inquire details.

Can I use a JCB credit card?

I'm sorry, but I can not use it.

PayPal settlement is possible if purchasing per mastering. If you can purchase V Preca at a convenience store, you can sign a premium plan.

I want to master in mono

Currently, when mastering a monaural sound source, it is converted to a stereo sound source and output. We are studying the function to output with mono sound source.

I also want to master songs with large dynamic range such as orchestra

It is possible to correspond to a certain extent by setting the target sound pressure to a lower value (such as -12 dB) by custom explosion and designating a song close to the image as a reference sound source.

Which sampling frequency can output?

You can choose from 44.1 kHz, 48 kHz, same as input.

Are there any recruitment for collaborative research?

Please contact us.

Is there a desktop version?

There is Windows version.

Please download from [ba_desktop_link]. The function is almost the same as the web service version, but the Windows version features that you can automatically upload the audio file written in the specified folder to AI Mastering.

Can I use it in browsers other than Chrome (Firefox, Internet Explorer, Safari, etc)?

Although it may work with other browsers, in principle, it supports only the latest Chrome. Those who use other browsers, sorry to trouble you, please install Chrome from here .

Is there a way to master for free at least twice a month?

If you introduce your friends with friends introduction program, you can earn mastering tickets that can be mastered for free. For details, please refer to the Referral Program . (login is needed)

Is there an affiliate program?

Yes. For details, please click here .


Block chain music streaming service How to earn tokens using Choon

I will show you how to earn tokens by distributing songs using Choon.

What is Choon?

Choon is a music streaming service and digital payment ecosystem with an Ethernet block chains designed to solve the task of the music industry.

In the music industry, there are issues such as a lot of margin taken from the artist to the listener until music arrives, transparency is insufficient, existing streaming service is not profitable, and the structure of the industry is old.

Choon solves these tasks in the block chain. Detail isWhite paperplease look at.

Register as a free member

HerePlease register as a member. * At the time of registration, please check "I am an artist who wants to upload music".

Upload sound source

Once you register as a member, you will be taken to the screen below. Click the "NEW TRACK" button to switch to the sound source upload screen.

Upload sound source from "Choose a file" of Audio.

Upload artwork images from Artwork Image. If there is no suitable image you can use the image without credit notationunsplashI think that it is good to use such as.

Enter meta information

Enter the following information.

Title: Title of the song

Artist Name: Artist name

Release Date: Release date

Description: Description

Tags: Tag

Decide playlist reward

Choon has a function to create playlists. "Playlist Split" is the proportion of remuneration divided into playlist creators when a song is played through a playlist. If you want rewards, you should consider a smaller value, if you want to increase the number of playback, I think that a larger value is better.

Enter compensation distribution information

Choon has the function of automatically distributing rewards among people when multiple people ("Collaborator") are involved in sound source production. Please enter the compensation distribution information from the Contract tab. In the case of one person, it is OK as default.


Save it and you're done.

How much can I actually earn?

actuallyPosted 8 songs to ChoonI tried to. After posting the first song on September 16, 2018, I earned 3,230 notes as of 2018/12/03. notes is the name of the token exchanged by Choon. 1 I do not know the value of notes yet.


I showed you how to earn tokens by distributing songs using Choon.

A smartphone app for increasing the volume & sound quality of videos taken with smartphones was released!

In this article, we will show you how to automatically master audio / video using AI Mastering 's smartphone version.

Install from the following App Store link


What is mastering?

Mastering is the task of adjusting the sound. By mastering, it makes it easier to hear, and powerful. Simply put, it makes a good sound. By realizing dynamic range, frequency characteristics, sound pressure (loudness) etc, realize it.

For music, professional, amateur, regardless of releasing the song, almost always, mastering is done. For movies, I think that mastering is probably done on commercially available movies and TV programs, but I think that many videos uploaded to YouTube are not mastered. When you listen to the sound of a TV or movie and the sound of YouTube, especially when you listen to the sound of YouTube, it is difficult to listen by the volume difference between the videos (especially difficult to listen to) Saya, time to adjust the volume), there is a difference.

For movies, the word mixing, mixing, MA may be more popular than mastering. What is the difference between these and mastering?

Simply put, it is an image that contains silence, mix and mastering in the MA. Recording -> MA (sound -> mix -> mastering) -> distribution. Since there is a possibility that opinions may differ more than any other person, I do not get deeply involved. ※ In music, the boundary between mix and mastering is ambiguous.

An era where you can shoot, edit, and upload videos with your smartphone

Already there are many useful video editing applications on the App Store. Combining these applications makes it easy to shoot, edit and upload movies with smartphones.

Furthermore, if you combine AI Mastering, you can also automatically master video and audio. You can expect to be able to create movies that make it easier to listen to sounds than ever.

* AI Mastering is an automatic mastering web service that has been provided for composers before. Since full-fledged mastering is possible than the smartphone version, if you are interested also, please do.

AI Mastering Features of the smartphone version

Automatic Mastering

Mastering automatically according to the input sound source.

noise reduction

Cut the noise that has been entered by recording.

YouTube Optimization

Adjust the volume to the optimum level according to YouTube's loudness normalization specification.

How to master audio of video on iPhone

Install the application

Install the application from the App Store.

Register as a free member

AI Mastering is a must-have application. Please sign up with social login by Google, Twitter, Github or by e-mail address and password. Accounts are shared with the smartphone version and web service version, so if you are already registered as a web service version, you can log in with the same ID.

I agree to the terms of service and register my email address

When membership registration is completed, we will transition to the screen below, so we will register your email address and confirm your terms of service.

By default, the mail address used at login is entered in the mail address input field. Please press OK as it is if there is no problem. If you do not enter it or change it, please press OK after input.

Upload a video

Upload videos saved in iPhone photos (photo library). Please press the new button on the screen after login. ※ The movie will not be published. It will be deleted from the old one in about 3 days.

Wait for mastering completion

Once the video is uploaded, the mastering process starts on the server. Eventually, please wait until video encoding is complete.

Download videos

Please click the download button and download the movie after mastering.

How to use is over.

How to use

Use as preprocessing (sound) of video editing

It is a method of using AI Mastering as the original material before video editing.

I think that when you record with iPhone's standard microphone, noise often comes in. Noise can be reduced by passing AI Mastering before video editing. Since the volume is also aligned, it will be easier to handle as a material for movie editing.

Use as a finishing (mastering) after video editing

Using AI Mastering as a finishing before uploading after attaching BGM or sound effect in movie editing.

Overall quality can be raised.


What is Loudness Normalization?

Loudness normalization is a function to make the volume of video constant. Since the volume difference between videos is reduced, you do not have to adjust the volume each time the user switches the video.

Loudness normalization is also introduced on YouTube. However, if the volume of the original movie is too small on YouTube, the volume after loudness normalization will remain small. Such videos are also frequent on top YouTuber videos.

With AI Mastering, optimize the volume when playing on YouTube so that it does not become smaller compared to other videos. You can prevent your video from losing power compared to other videos on YouTube.

Loudness normalization has already been introduced on YouTube. Is there any meaning to increase sound pressure now?

It is necessary to ensure the minimum sound pressure according to the level of the loudness normalization, but there is no need to pursue further sound pressure.

However, as a result of analyzing YouTube's video sound, there are not many videos on YouTube that have reached the minimum sound pressure. Easily securing the minimum sound pressure with AI Mastering makes it sound comparable to the surrounding sound. ※ The results of the survey will be reported separately.

Since sound pressure and sound quality are in a tradeoff, it is best to aim at the very close sound pressure at which volume reduction by loudness normalization occurs. Therefore, the sound pressure setting of the smartphone version of AI Mastering is set lower than the web service version.

Using AI Mastering, you can raise the sound pressure without thinking about tradeoffs with sound quality at minimum sound pressure.

Does the volume rise so much that the sound breaks?

AI Mastering has a mechanism to prevent sound cracking called a limiter. The limiter of AI Mastering raises the volume while checking whether the sound is broken, so it may occasionally crack sound, but in most cases it will not sound crack.

However, please note that there may be a sound crack due to re-encoding. Re-encoding is to decode a video encoded in mp4 format etc. once and encode again. As a result, the form of the speech waveform changes and the peak jumps out, which may distort the sound.

Re-encoding is done when editing video or uploading to a video site. AI Mastering gives some margin to the peak in order to prevent sound cracking due to re-encoding, but how much peak will pop out by re-encoding greatly depends on re-encoding settings and the original sound source.

If you break the sound after uploading the movie, if you tell us the site you uploaded (YouTube etc.), we will respond.

What is the difference with other video editing applications?

The difference between AI Mastering and other video editing applications is where AI Mastering specializes in sound. Because I can not edit video instead of specializing in sound, I think that it is good to use it together with other video editing application.

It is a comparison table with movie editing application that can be used on iPhone.

 Sound pressure bombardment kunPremiere RushPremiere ClipiMovieInShotVivaVideo
Supported environmentiPhone, PCiPhone, iPad, PC (Premiere)iPhone, iPad, Android, PC (Premiere)iPhone, iPad, PCiPhone, iPad, AndroidiPhone, iPad, Android
Automatic Masteringoxxxxx
Overall volume adjustment (YouTube optimization)oxxxxx
Dynamic range adjustmentoo (sound balance)xxxx
noise reductionooxxxx
Echo reductionxoxxxx
Speech emphasisxoxxxx
Adjust volume between clipsN / Ao (Automatic volume)o (Smart volume)xxx
DuckingN / Aoxxxx
Movie editingxooooo
Selling agencyOur companyAdobeAdobeAppleInstaShot Inc.QuVideo Inc.

* Applications that can be used on PC are also included, but they are compared by the function of the smart version. x means that the function could not be confirmed on the surface. It may be implicitly implemented.

Will you be charged?

The smartphone version is free. A portion of the Web service version is charged. For details of the Web service version, please see AI Mastering .

What is the difference from the web service version?

The application version is specialized in movies with simple functions. The web service version is generic and can be used for both music production and motion picture production use. Since user accounts are shared, you can log in to both the application version and the web service version with the same ID. It is also possible to download videos uploaded with the application version from the web service version. Prices are free for all application versions, some for free on the web service version and some are charged.

Please see AI Mastering for details of the web service version.

Is there an iPad version?

It is under development.

Is there an Android version?

It is under development.

Is the sound source uploaded to the server public?

It will not be published.

Will the sound source uploaded to the server be deleted?

It will be deleted in about 3 days from the old one.

Does communication occur?

Because it processes video on server, communication volume is large. Please use it in Wifi environment.

Is there a version that processes locally only without communicating?

There is no present condition. It is under consideration.


※ The information in this article is as of 12/02/2018, may differ from the latest information. For the latest information, please see AI Mastering or the App Store.

How to participate in AI Mastering's affiliate program

In this article, we will show you how to join AI Mastering 's affiliate program for free for affiliate (affiliate marketer).

What is AI Mastering?

AI Mastering is an automatic online audio mastering service.

Mastering is the task of adjusting the sound. It is necessary for music production and movie production. Mastering makes the sound easier to hear, the impressions come out and the impression gets better.

With AI Mastering, mastering can be done with one touch.

Affiliate Program Overview

When you participate in AI Mastering's affiliate program, you can receive affiliate fee when introducing AI Mastering's new member registration and new premium plan registration.

Participation in the affiliate program is free.

Affiliate is provided by itself without using ASP (Affiliate Service Provider), so it is characterized by no margin by ASP.

Affiliate Program Rewards

Affiliate fees of $ 0.05 will be incurred when introducing new member registration and $ 15 when introducing new premium plan registration. The compensation generation condition is the same as the Referral Program . For details, please visit there. (login is needed)

How to participate in affiliate program

Free membership with AI Mastering

Affiliate program of AI Mastering does not use affiliate network, so we manage affiliate fee with account of AI Mastering. Please do a free member registration from here .

If you do not know how to register as a member,This articlePlease also refer to it.

Apply for participation by e-mail

Please send a mail with the following contents to aimasteringcom@gmail.com. To participate, it is necessary to agree to Affiliate Program Terms of Service .

For PayPal account, please see here .

 To: aimasteringcom@gmail.com
Subject: Sound pressure bomb kun affiliation program Application details:

I have read and agree to https: //aimastering.com/affiliate_terms_of_service/,
I will apply for participation in the affiliate program of sound pressure explosion kun.
Below is application information.

E-mail address registered for sound pressure bomb kun: (Please input)
Remuneration receipt account type: PayPal account / Japanese bank account (Please choose one or the other)

## When choosing PayPal account E-mail address of PayPal account or PayPal.Me's address: (Please enter)

## When choosing Japanese bank account Bank name: (please enter)
Branch name: (Please enter)
Branch code: (Please enter)
Account type: (Please enter)
Account number: (Please enter)
Name: (Please enter)

that's all, thank you very much. 

※ At the time of sending e-mail, it is considered to have agreed to Affiliate Program Terms of Service.

Wait for approval

After you activate your affiliate program with the AI ​​Mastering account corresponding to the email address you sent, I will reply by e-mail. Please wait a moment.

Acquire the introduction URL for affiliate and put it on blog etc.

Acquire the introduction URL for affiliate from affiliate program , and paste it on blog etc. For referrals, materials for affiliate programs described below can also be used.

Review report

You can check the report from the Affiliate Program . However, it will not be displayed unless there is at least one referral.

Material for affiliate program

Materials listed in the following articles can be freely processed and used for affiliate purposes.

Material collection for affiliate program


What is the difference between introducing friends?

Your friends referral and payment terms are the same, but the content of remuneration is different. Participants in the affiliate program can receive affiliate fees in addition to the referring fee for referring friends when the compensation granting conditions are satisfied. Please refer to the Referral Program for details of compensation generation conditions and so on. (login is needed)

Will the reward of the premium plan be received every month after the first time if the user continues to use it?

I can not receive it. Remuneration will be given only during introduction.


We showed you how to participate in AI Mastering's affiliate program.