Audio Visualizer using Emotion
Content
Audio Equalizer
TABLE OF CONTENTS
The current report will describe and provide an informative breakdown of an audio equalizer
created to adjustable settings per any user preferences using LabVIEW. The final project for
Project V, Digital Signal Processing System Design, was to apply basic theory and applications of
modern digital signal processing, to learn basic theory of real-time digital signal processing, and
to develop ability to implement and simulate digital signal processing algorithms using MATLAB
and on real-time DSP platform. Within the project, we must demonstrate that these course
objectives have been learned:
Understand basic concepts of digital signal processing theories and techniques.
Develop basic understanding of real-time digital signal processing.
Develop ability to implement digital signal processing algorithms in Matlab and LabVIEW.
Develop ability to implement digital signal processing algorithms on real-time DSP
platform.
Background Information
1
An audio equalizer is a device commonly used in sound recording and reproduction to alter the
frequency response of an audio system using linear filters. 1 Equalization is can be broken into
three parts to better understand the process. The three parts are signal acquisition (input),
signal processing, and the output of the modified signal. Filters are the main component to the
equalization process. These filters can be used for noise suppression, signal enhancement,
removal or attenuation of a specific frequency. The signal processing section of the filters can be
. Audio equalizers are typically constructed in a parallel-circuit manner, where the lowpass,
bandpass, and highpass are connected in this configuration. The filters are set to specified
frequencies based on typical frequencies that can be heard by the average human ear.
Frequencies affected can be analyzed visually by implementing a spectrum analyzer. This
element allows the user to see and measure the effects as the controls are moved to a particular
preference. Such equipment to replicate these specific frequencies can be relatively expensive.
This can be corrected by using software, such as LabVIEW. This can minimize any size
restrictions and makes it portable for easy transport.
Signal Acquisition
The design of the audio equalizer allows a user to connect audio files via the 3.5 mm jack. This
was chosen to allow compatibility with most audio input devices. The myDAQ is used to acquire
the input signal. The myDAQ is a low-cost data acquisition (DAQ) device that gives students the
ability to measure and analyze live signals. National Instruments myDAQ is compact and
portable so students can extend hands-on learning outside of the lab environment using
industry-standard tools and methods.2 Once the myDAQ is connected to LabVIEW, we had to
program the correct to allow the audio to feed into our project’s VI. We had to set the sample
rate and voltage specifications to allow the audio files to play without lag or error. For example,
the voltage specifications must be set to ±2V for the left and right channel for the input and
output. I believe this is just a standard parameter for the myDAQ because once the voltage
parameters are exceed, an error dialog box will display and stop the VI from continuing. In the
diagrams below, we will show how the signal acquisition part of the VI looks:
1 Wikipedia
2
http://sine.ni.com/np/app/main/p/ap/academic/lang/en/pg/1/sn/n17:academic,n21:16781/fmid/63
53/
2
Settings are adjusted to
accommodate audio from
the 3.5mm jack of the left
and right channels.
Filters and Signal Processing
Three filters are created based on common usage in the audio market. These filters are the
bass, mid, and treble. Each filter is set to attenuate a specific frequency to enhance the signal to
a user preference. The bass filter is called a lowpass filter that removes frequencies above the
512 hertz. In this particular project, we used a Butterworth IIR filter. IIR filter stands for Infinite
Impulse Response which is used when filters are needed for “continuous-time” processing. The
second filter is for the mid-tones. A bandpass is created to handle a range of frequencies,
instead of having one cutoff frequency. The mid-tones are primarily utilized to enhance vocals
and certain intrinsic sounds that are not typically in the lower frequencies. These frequencies
range between 513 and 5000 hertz. Lastly, the treble is created to enhance the higher
frequencies above 5k hertz. Each filter is connected to a control element that allows user to
control the amount of output. These control elements are multiplied to have a large multiplied
that is noticeable in the audio files while being played. After the each filter is altered, then each
adjustment is added together to form the final output that is heard though the left and right
channels. An additional feature that is apparent on the front panel is the left and right balance.
This allows the user to adjust the output to play more to one side. For example, if one is seating
closer to the left speaker, he or she can shift the toggle to have more audio to the left verses the
right.
3
Each filter is calculated separately and then
summed together to form the final output
that is heard through the speakers with the
desired adjustments from the user.
Filter Configurations in LabVIEW
Lowpass Filter: Bass (Left), Bandpass Filter: Mid-tone (Right), & Highpass Filter:
Treble (Below)
4
Signal Output and Display
The signal output is displayed in three different methods, Sound, Spectral Analysis and
Frequency Bands. The myDAQ used again to play the altered audio signal through generating
output signals for the left and right channel speakers. The best way to enjoy music is through
sound. The settings to generate an output signal with the myDAQ are very similar to the input
settings to acquire the signal. As you can see above in the block diagram, the final blue data line
is connected to myDAQ and the spectral measurement element. This allows one to visualize the
data as the sound is being played. The spectral measurement tool performs FFT-based spectral
measurements, such as the averaged magnitude spectrum, power spectrum, and phase
spectrum on a signal. Thirdly, a frequency band was created to visually display the varying
frequencies from the audio files. In order to develop the bands, we have to covert the input data
into a sine wave. The sine wave is the scaled down into an array which called be later converted
into decibel units. The varying decibels from the audio files are grouped into 5 different arrays
depending of the frequency level. This is displayed on the front panel as five columns consisting
of Boolean indicators. As each frequency range increases or decreases, the indicators match the
variance from green, yellow, and red.
Front Panel Display
Week-to-Week Task List
Wee
k
1
Task(s)
Gather ideas and complete research about audio equalizers.
Brainstorm to determine best method to implement the
researched data into LabVIEW. Develop outline for report
5
2
3
4
and power point presentation.
Build project within LabVIEW.
Troubleshot any problems encountered during the build.
Include screenshots and begin typing the report for the
project.
Practice for the presentation, possibly with notecards. Check
that objectives and goals set have been achieved. Make
corrections related to format or usage in the report and
presentation.
Conclusion
The audio equalizer is common in different real world applications. LabVIEW can be used to
reduce cost provide the same effects as you higher end audio equipment. Challenges did occur
throughout the process. Many hours were dedicated to research to understand the elements
required to create a frequency band within the software. Also, we had difficulty developing the
correct calculations to have the frequency vary as the audio files were playing. The filters and
controls for the filters had to be constantly changed to accommodate the small voltage output of
±2V. Overall, this was an interesting because it required us to use many engineering design
standards and learn the extensive elements within LabVIEW to achieve the completion of our
final project.
References
I.
Welch, Thad, Wright, Cameron, Morrow, Michael. Real-Time Digital Signal
Processing from MATLAB to C with the TMS3206x DSPs. Boca Raton,FL: CRC
Press.2012
6
TABLE OF CONTENTS
The current report will describe and provide an informative breakdown of an audio equalizer
created to adjustable settings per any user preferences using LabVIEW. The final project for
Project V, Digital Signal Processing System Design, was to apply basic theory and applications of
modern digital signal processing, to learn basic theory of real-time digital signal processing, and
to develop ability to implement and simulate digital signal processing algorithms using MATLAB
and on real-time DSP platform. Within the project, we must demonstrate that these course
objectives have been learned:
Understand basic concepts of digital signal processing theories and techniques.
Develop basic understanding of real-time digital signal processing.
Develop ability to implement digital signal processing algorithms in Matlab and LabVIEW.
Develop ability to implement digital signal processing algorithms on real-time DSP
platform.
Background Information
1
An audio equalizer is a device commonly used in sound recording and reproduction to alter the
frequency response of an audio system using linear filters. 1 Equalization is can be broken into
three parts to better understand the process. The three parts are signal acquisition (input),
signal processing, and the output of the modified signal. Filters are the main component to the
equalization process. These filters can be used for noise suppression, signal enhancement,
removal or attenuation of a specific frequency. The signal processing section of the filters can be
. Audio equalizers are typically constructed in a parallel-circuit manner, where the lowpass,
bandpass, and highpass are connected in this configuration. The filters are set to specified
frequencies based on typical frequencies that can be heard by the average human ear.
Frequencies affected can be analyzed visually by implementing a spectrum analyzer. This
element allows the user to see and measure the effects as the controls are moved to a particular
preference. Such equipment to replicate these specific frequencies can be relatively expensive.
This can be corrected by using software, such as LabVIEW. This can minimize any size
restrictions and makes it portable for easy transport.
Signal Acquisition
The design of the audio equalizer allows a user to connect audio files via the 3.5 mm jack. This
was chosen to allow compatibility with most audio input devices. The myDAQ is used to acquire
the input signal. The myDAQ is a low-cost data acquisition (DAQ) device that gives students the
ability to measure and analyze live signals. National Instruments myDAQ is compact and
portable so students can extend hands-on learning outside of the lab environment using
industry-standard tools and methods.2 Once the myDAQ is connected to LabVIEW, we had to
program the correct to allow the audio to feed into our project’s VI. We had to set the sample
rate and voltage specifications to allow the audio files to play without lag or error. For example,
the voltage specifications must be set to ±2V for the left and right channel for the input and
output. I believe this is just a standard parameter for the myDAQ because once the voltage
parameters are exceed, an error dialog box will display and stop the VI from continuing. In the
diagrams below, we will show how the signal acquisition part of the VI looks:
1 Wikipedia
2
http://sine.ni.com/np/app/main/p/ap/academic/lang/en/pg/1/sn/n17:academic,n21:16781/fmid/63
53/
2
Settings are adjusted to
accommodate audio from
the 3.5mm jack of the left
and right channels.
Filters and Signal Processing
Three filters are created based on common usage in the audio market. These filters are the
bass, mid, and treble. Each filter is set to attenuate a specific frequency to enhance the signal to
a user preference. The bass filter is called a lowpass filter that removes frequencies above the
512 hertz. In this particular project, we used a Butterworth IIR filter. IIR filter stands for Infinite
Impulse Response which is used when filters are needed for “continuous-time” processing. The
second filter is for the mid-tones. A bandpass is created to handle a range of frequencies,
instead of having one cutoff frequency. The mid-tones are primarily utilized to enhance vocals
and certain intrinsic sounds that are not typically in the lower frequencies. These frequencies
range between 513 and 5000 hertz. Lastly, the treble is created to enhance the higher
frequencies above 5k hertz. Each filter is connected to a control element that allows user to
control the amount of output. These control elements are multiplied to have a large multiplied
that is noticeable in the audio files while being played. After the each filter is altered, then each
adjustment is added together to form the final output that is heard though the left and right
channels. An additional feature that is apparent on the front panel is the left and right balance.
This allows the user to adjust the output to play more to one side. For example, if one is seating
closer to the left speaker, he or she can shift the toggle to have more audio to the left verses the
right.
3
Each filter is calculated separately and then
summed together to form the final output
that is heard through the speakers with the
desired adjustments from the user.
Filter Configurations in LabVIEW
Lowpass Filter: Bass (Left), Bandpass Filter: Mid-tone (Right), & Highpass Filter:
Treble (Below)
4
Signal Output and Display
The signal output is displayed in three different methods, Sound, Spectral Analysis and
Frequency Bands. The myDAQ used again to play the altered audio signal through generating
output signals for the left and right channel speakers. The best way to enjoy music is through
sound. The settings to generate an output signal with the myDAQ are very similar to the input
settings to acquire the signal. As you can see above in the block diagram, the final blue data line
is connected to myDAQ and the spectral measurement element. This allows one to visualize the
data as the sound is being played. The spectral measurement tool performs FFT-based spectral
measurements, such as the averaged magnitude spectrum, power spectrum, and phase
spectrum on a signal. Thirdly, a frequency band was created to visually display the varying
frequencies from the audio files. In order to develop the bands, we have to covert the input data
into a sine wave. The sine wave is the scaled down into an array which called be later converted
into decibel units. The varying decibels from the audio files are grouped into 5 different arrays
depending of the frequency level. This is displayed on the front panel as five columns consisting
of Boolean indicators. As each frequency range increases or decreases, the indicators match the
variance from green, yellow, and red.
Front Panel Display
Week-to-Week Task List
Wee
k
1
Task(s)
Gather ideas and complete research about audio equalizers.
Brainstorm to determine best method to implement the
researched data into LabVIEW. Develop outline for report
5
2
3
4
and power point presentation.
Build project within LabVIEW.
Troubleshot any problems encountered during the build.
Include screenshots and begin typing the report for the
project.
Practice for the presentation, possibly with notecards. Check
that objectives and goals set have been achieved. Make
corrections related to format or usage in the report and
presentation.
Conclusion
The audio equalizer is common in different real world applications. LabVIEW can be used to
reduce cost provide the same effects as you higher end audio equipment. Challenges did occur
throughout the process. Many hours were dedicated to research to understand the elements
required to create a frequency band within the software. Also, we had difficulty developing the
correct calculations to have the frequency vary as the audio files were playing. The filters and
controls for the filters had to be constantly changed to accommodate the small voltage output of
±2V. Overall, this was an interesting because it required us to use many engineering design
standards and learn the extensive elements within LabVIEW to achieve the completion of our
final project.
References
I.
Welch, Thad, Wright, Cameron, Morrow, Michael. Real-Time Digital Signal
Processing from MATLAB to C with the TMS3206x DSPs. Boca Raton,FL: CRC
Press.2012
6
