Pitch Correction Tutorial in Adobe Audition and Audacity-Tuning Recorded Audio Tracks

This is a tutorial on how to implement pitch correction in Adobe Audition. Bear in mind that re-recording is still the best way to correct out-of tune instruments.

It is why you need to carefully check the tuning of your instruments with a digital tuner before doing any recording. In case you really need to implement pitch correction feature, below are the following steps:

The Complete Steps in Adobe Audition

Step1. Open the audio track in Adobe Audition Edit View.

Tips: If you are using pitch correction to aim for best recording quality results, it is good to use a high resolution audio. High resolution audio should be better than CD audio or MP3. In this case, 24 bit and 44.1 KHz wav file is the recommended format. If you asked why, it’s because you are doing digital manipulations to the wav file. High resolution audio obviously contains more digital audio information (in terms of digital bits) as compared to ordinary CD audio/mp3.

Step2. Go to File – Save as. You need to save a copy of your wav file as a backup. Assign a different name. This is important in case you need to revert to the original wave unedited with pitch correction.

Step3. In Adobe audition, notes are detected in terms of chord names. Chord names are A-A#-B-C-C#-D-D#-E-F-F#-G-G#-A. For example say you have a 3.45 seconds bass guitar riff or passage that should be in a G major chord. You can check whether the recorded passage is actually a G major chord. To do this, highlight the 3.45 seconds passage.

Step4. Go to Analyze – Show Frequency Analysis. Set the FFT size to 65536/Blackmann-Harris settings. Then click “Scan”. Below is the screenshot of the results:

It shows that the 3.45 seconds audio passage is actually having G chord sound. No need to use pitch correction to this passage because it is “in-tune”. You have just learned how to verify the pitch/tuning in Adobe audition edit view using the built-in frequency analysis tool.

Step5. Supposing the next passage (after the 3.45 seconds) should be an E-minor chord. You can select that passage and repeat the above steps to analyze the chords. Screenshot:

It shows that the recorded passage is “out of tune”. It is because you are expecting it to be “E” –chord sound but the recorded passage is having a “D#” sound.

Step6. To correct the pitch of the out of tune recording passage; go to Effects – Time/Pitch – Pitch Correction.

Step7. Go to “Automatic” Tab and configure as follows:

Scale: Minor
Key: E
Attack: Middle (between slow and fast)
Sensitivity: 110
FFT size: 4096
Calibration: 440Hz

It should like this one below:

Finally, click OK.

Step8. Verify the pitch again of the corrected audio passage using frequency analysis tool using the procedure illustration in Step4.

It now shows that the pitch of the out-of-tune wave is now having an E minor chord sound (as pitch correction from D# minor to E-minor)

You can even extend this pitch correction technique to not only tuning chords in a passage but to single notes. You just need to zoom the audio waveform as high as possible so that you can analyze on a note per note basis. Same concept/principle can be implemented as illustrated in this tutorial.

However, pitch correction is ideal for editing a small portion of the wave. It is not recommended to do major pitch correction editing for the entire recorded tracks because this can seriously affect the resulting sound quality.

Implementing Pitch Correction in Audacity

One of the challenging audio editing tasks is to change the pitch of the music. This is particularly useful if a certain singer needs a change of pitch or a key signature to fit his/her vocal range.

If the pitch is high for the singer, the best suggestion would be to lower down the key signature of the song, this is done by lowering the pitch of the entire music notes. So if the current key signature is G major, the singer might need to transpose it down to F major or E major, depending on the choice.

The same case if a certain singer finds out that the current key is too low for the vocal range and wants to increase the pitch of the background music.

Note: Changing pitch does not necessarily change the key signature of the music at all times. Example of this is when you transpose the pitch to one octave. The key signature would stay the same. But changing key signature would always change the pitch of the music. You need to bear this in mind.

It is interesting to know that this feature is available in open source audio editor like Audacity. Not only changing the pitch but normalizing the audio wave is also important. This is the process of maximizing the volume of the audio without the use of compression or limiting.

You need to follow the steps below:

Step1.) Launch Audacity.
Step2.) Open your Audio project that you need to change pitch. (File – Open). The source audio can be an MP3 file. But take note that if you want better quality, you might want to convert from the highest resolution format available. So if the music to be transposed has the following available formats:

a.) 24-bit/48KHz
b.) 16-bit/44.1KHz CD audio
c.) 320kbps MP3
d.) 128kbps MP3

Then use the 24-bit/48KHz format because its available and it has the highest resolution. Make sure you make some backup to this file because pitch transposition is a destructive form of audio editing. Put the backup in a different place, e.g. in a external hard drive.

Also bear in mind that converting from low to high resolution does not improve the quality of your audio to be transposed. So if you only have 128Kbps MP3 format and you convert it to 16-bit/44.1KHz or even 24-bit/48KHz, then you will get no benefit in terms of sound quality. It is because the source audio is already in low resolution.

Step3.) Now go to Effect – Change Pitch. The Change pitch tool automatically detects the current Key signature of the music, for example in the screenshot below:

The current pitch of the music is G. Supposing the singer wants to bring down the pitch by a half step. So the new pitch would be F#/Gb.

To change this, simply select the radio button “down”, and set “To:” F#/Gb. This setting would transpose the entire pitch of the music by a half step. This is the new setting as shown below:

Otherwise if you want a higher pitch setting. Supposing from G key to A. Then select “up” and then select A pitch.

Note: You do not need to change or edit the following fields:
a.) Semitones
b.) Frequency from and to.
c.) Percent change.

Leave all the above mentioned fields by default.

Step4.) You can even listen to the sample preview by clicking the preview button. This is highly recommended so that you will not be repeating the pitch process over again that can be time consuming. Let the singer practice the preview pitch to make sure it is already final. If all are set, click OK.

Step5.) You can save the file. The most recommended format for saving the edited file is in 24-bits/44.1KHz WAV. First, you need to save the project. Go to File — > Save Project As then type any desired name of the project and click save.

Then go to File — > Export, type the name of the edited file (do not overwrite the original audio source or your backup!). Then in the “Save as type”, select “Other uncompressed files”. Click “Options” and then under “Header” select WAV (Microsoft). Then under “Encoding”, select “Signed 24-bit PCM”. Click OK and save. You can optionally enter metadata. You can skip this by clicking OK. Close the file and exit Audacity.

Creating a copy of transposed audio or distribution

By following steps in the previous section, you now have a high resolution 24-bit copy of the transposed music. Your objective is to create copies that you can distribute to other medium such as CD or MP3. To do this, follow the steps below:

Step1.) In Audacity, go to File — > Open. You need to load the edited file (24-bit audio).

Step2.) Check first if the audio volume has been maximized. Go to Effect — > Normalize. Check all available options and then set the maximum amplitude to -0.2dB.

Step3.) Click OK. The transposed music is now louder than before.

Step4.) Export the audio again as 24-bits following step 5 in the previous section.

Step5.) Download this tool.

Step6.) Using Voxengo R8brain, you need to implement dithering and sample rate conversion (SRC). This will convert the 24-bit audio to 16-bits/44.1KHz. SRC is only needed when the original audio sample rate is not equal to 44.1KHz.

Step7.) What you have after R8brain conversion is a 16-bit/44.1KHz WAV. If you like to convert it to MP3, to export it as MP3. If it looks for a LAME encoder. Download it here.

The rest will be self-explanatory.

Recommendations for Best Results

1.) Always use the highest resolution audio format as possible when dong audio editing.
2.) Don’t use MP3 or lower WAV resolution such as 16-bit.
3.) If you can re-record the out-of tune performance, then much better. Re-recording is always preferred than serious digital editing. Why? Do you know that heavy digital editing such as pitch correction can introduce some unwanted side effects? These side effects would be audible to your listeners.
4.) The quality of the edited result strongly depends on the quality or algorithm of the pitch correction software.
5.) If you don’t have Adobe Audition, there are lots of free software that allows you to edit the pitch of the waveform. For example, you can use Audacity and its free plugins.

For more information, read the additional/related articles of this post below.

Content last updated on October 13, 2012