Presentation is loading. Please wait.

Presentation is loading. Please wait.

Speech/Audio Signal Processing in MATLAB/Simulink

Similar presentations


Presentation on theme: "Speech/Audio Signal Processing in MATLAB/Simulink"— Presentation transcript:

1 Speech/Audio Signal Processing in MATLAB/Simulink
2018/11/17 2006 Speech/Audio Signal Processing in MATLAB/Simulink Speech/Audio Signal Processing in MATLAB/Simulink J.-S. Roger Jang (張智星) CS Dept, Tsing-Hua Univ, Taiwan (清華大學 資訊系) ... In this talk, we are going to apply two neural network controller design techniques to fuzzy controllers, and construct the so-called on-line adaptive neuro-fuzzy controllers for nonlinear control systems. We are going to use MATLAB, SIMULINK and Handle Graphics to demonstrate the concept. So you can also get a preview of some of the features of the Fuzzy Logic Toolbox, or FLT, version 2.

2 Outline Wave file manipulation Time-domain processing
2018/11/17 Outline Wave file manipulation Reading, writing, recording ... Time-domain processing Delay, filtering, sptools … Frequency-domain processing Spectrogram Pitch determination Auto-correlation, SIFT, AMDF, HPS ... Others Formant estimation, speech coding Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

3 Toolbox/Blockset Used
2018/11/17 Toolbox/Blockset Used MATLAB Simulink Signal Processing Toolbox DSP Blockset Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

4 2018/11/17 MATLAB Primer Before you start, you need to get familiar with MATLAB. Please read “MATLAB Primer” at the following page: Exercise: Please plot two curves y=sin(2*t) and y=cos(3*t) in the same figure. Please plot x vs. y where x=sin(2*t) and y=cos(3*t). 2018/11/17

5 To Read a Wave File To read a MS .wav file (PCM format only): wavread
2018/11/17 To Read a Wave File To read a MS .wav file (PCM format only): wavread y = wavread(file) […] = wavread(file, [n1, n2]) [y, fs, nbits, opts] = wavread(file) […] = wavread(file, n) [y, fs, nbits] = wavread(file) If the wav file is stereo, y will be a two-column matrix. Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

6 To Read a Wav File Example (wavRead01.m): Exercise:
2018/11/17 To Read a Wav File Example (wavRead01.m): [y, fs] = wavread('singapore.wav'); plot((1:length(y))/fs, y); xlabel('Time in seconds'); ylabel('Amplitude'); Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. Exercise: Plot the waveform of “rrrrr.wav”. Use MATLAB’s “zoom” button to find the consecutive curling “R” occurs. Plot the two-channel waveform in “flanger.wav”. 2018/11/17

7 Solution to the Previous Exercise
2018/11/17 Solution to the Previous Exercise wavRead02.m: [y, fs] = wavread(‘flanger.wav’); subplot(2,1,1), plot((1:length(y))/fs, y(:,1)); subplot(2,1,2), plot((1:length(y))/fs, y(:,2)); 2018/11/17

8 2018/11/17 To Play Wav Files To play sound using Windows audio output device: wavplay, sound, soundsc wavplay(y, fs) wavplay(y, fs, ‘async’): non-blocking call wavplay(y, fs, ‘sync’): blocking call sound(y, fs) soundsc(…): autoscale the sound Example (wavPlay01.m): [y, fs] = wavread(‘rrrrr.wav’); wavplay(y, fs); Exercise: Follow the example to play “flanger.wav”. Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

9 To Read/Play Using DSP Blocks
2018/11/17 To Read/Play Using DSP Blocks To read/play sound using DSP Blockset: DSP Blockset/DSP Sources/From Wave File DSP Blockset/DSP Sinks/To Wave Device Example: Exercise: Create a model as shown above. The file name is “waveio.mdl”. Frame-based operation! 2018/11/17

10 Solution Solution to the previous exercise: slWavFilePlay01.mdl
2018/11/17 Solution Solution to the previous exercise: slWavFilePlay01.mdl 2018/11/17

11 To Write a Wave File To write MS wave files: wavwrite
2018/11/17 To Write a Wave File To write MS wave files: wavwrite wavwrite(y, fs, nbits, wavefile) “nbits” must be 8 or 16. “y” must have two columns for stereo data. Amplitude values outside [-1,1] are clipped. Example (wavWrite01.m): [y, fs] = wavread(‘rrrrr.wav’); wavwrite(y, fs*1.2, 8, ‘testout.wav’); !start testout.wav Exercise: Try out the above example. Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

12 To Record a Wave File To record wave files: Example: Exercise:
2018/11/17 To Record a Wave File To record wave files: 1. Use the recording utility under WinXP. 2. Use “wavrecord” under MATLAB. 3. Use “From Wave Device” under Simulink, under “DSP Blocksets/Platform Specific IO/Windows (Win32)” Example: 1. Go ahead and try WinXP recording utility! 2. Try “wavRecord01.m” 3. Try “slWavFileRecord01.mdl” Exercise: Try out the above examples. Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

13 Time-Domain Speech Signals
2018/11/17 Time-Domain Speech Signals A typical time-domain plot of speech signals: Amplitude: volume or intensity Frequency: pitch To generate the plot, try “plotwave.m”. 2018/11/17

14 Changing Wave Playback Param.
2018/11/17 Changing Wave Playback Param. To control the play of a sound: Normal: wavplay(y, fs) High volume: wavplay(2*y, fs) Low volume: wavplay(0.5*y, fs) High pitch (and faster): wavplay(y, 1.2*fs) Low pitch (and slower): wavplay(y, 0.8*fs) Exercise: Try “wavPlay01.m” and trace the code. Create “wavPlay02.m” such that you can record your own voice on the fly. Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

15 Time-Domain Signal Processing
2018/11/17 Time-Domain Signal Processing Take-home exrecise: How to get a high pitch with the same time span? Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

16 2018/11/17 Synthetic Sounds Use a sine wave generator (under DSP blocksets) to produce sounds Single frequency: Multiple frequencies: Amplitude modulation: Exercise: Create the above models. Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

17 Solution Solution to the previous exercise: sineSource01 sineSource02
2018/11/17 Solution Solution to the previous exercise: sineSource01 sineSource02 sineSource03 2018/11/17

18 Delay in Speech/Audio What is a delay in a signal?
2018/11/17 Delay in Speech/Audio What is a delay in a signal? y(n) --> y(n-k) What effects can delay generate? Echo Reverberation Chorus Flanging Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

19 Single Delay in Audio Signal
2018/11/17 Single Delay in Audio Signal Block diagram: z -k Input a Output u(n) y(n) = u(n) + a*u(n-k) Simulink model: The file is “delay1.mdl”. Exercise: Create the above model. 2018/11/17

20 Multiple Delay in Audio Signal
2018/11/17 Multiple Delay in Audio Signal How to create “karaoke” effects: a z -k Input Output y(n) u(n) 2 3 y(n) = u(n) + a u(n-k) + a u(n-2k) + a u(n-3k) ... Delay2.mdl Simulink model: 2018/11/17

21 Multiple Delay in Audio Signal
2018/11/17 Multiple Delay in Audio Signal Parameter values: Feedback gain a < 1 Actual delay time = k/fs Exercise: Create the above model and change some parameters to see their effects. Modify the model to take microphone input (so you can start singing karaoke now!) Use a “configurable subsystem” to include all possible input files and the microphone. (See next page.) Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

22 Multiple Delay in Audio Signal
2018/11/17 Multiple Delay in Audio Signal How to use “configurable subsystem” block? 1. Create a library (say, wavinput.mdl) 2. Get a block of “configurable subsystem” 3. Fill the dialog box with the library name Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

23 Audio Flanging Flanging sound: Simulink demo:
2018/11/17 Audio Flanging Flanging sound: A sound similar to the sound of a jet plane flying overhead, or a "whooshing" sound “Pitch modulation” due to a variable delay Simulink demo: dspafxf.mdl (all platforms) dspafxf_nt.mdl (for 95/98/NT) Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

24 Original spectrogram: Modified spectrogram:
2018/11/17 Audio Flanging Simulink model: Original spectrogram: Modified spectrogram: Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

25 Signal Processing Using sptool
2018/11/17 Signal Processing Using sptool To invoke sptool, type “sptool”. Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

26 Speech Production How is speech produced? Analogy to System Theory:
2018/11/17 Speech Production How is speech produced? Speech is produced when air is forced from the lungs through the vocal cords (glottis) and along the vocal tract. Analogy to System Theory: Input: air forced into the vocal cords Output: media vibration System (or filter): vocal tract Pitch frequency: frequency of the input Formant frequency: resonant frequency Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

27 Source Filter Model of Speech
2018/11/17 Source Filter Model of Speech The source-filter model of speech production: Speech is split into a rapidly varying excitation signal and a slowly varying filter. The envelope of the power spectra contains the vocal tract information. Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. Two important characteristics of the model are fundamental (pitch) frequency (f0) and formants (F1, F2, F3, …) 2018/11/17

28 Frame Analysis of Speech Signal
2018/11/17 Frame Analysis of Speech Signal Speech wave form : Zoom in Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. Overlap Frame 2018/11/17

29 2018/11/17 Spectrogram Spectrogram (specgram.m) displays short-time frequency contents: Wave form : Spectrogram : 2018/11/17

30 Real-time Spectrogram
2018/11/17 Real-time Spectrogram Try “dspstfft_win32”: Spectrum: Spectrogram: 2018/11/17

31 Pitch and Formants Pitch and formants can be defined visually:
2018/11/17 Pitch and Formants Pitch and formants can be defined visually: Pitch period = 1/f0 First formant F1 Second formant F2 2018/11/17

32 Spectrogram Reading Spectrogram Reading
2018/11/17 Spectrogram Reading Spectrogram Reading Waveform: Spectrogram: 2018/11/17 “compute”

33 Pitch Determination Algorithms
2018/11/17 Pitch Determination Algorithms Time-domain: Auto-correlation AMDF (Average Magnitude Difference Function) Gold-Rabiner algorithm (1969) Frequency-domain: Cepstrum (Noll 1964) Harmonic product spectrum (Schroeder 1968) Others: SIFT (Simple inverse filter tracking) Maximum likelihood Neural network approach Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

34 Autocorrelation of Each Frame
2018/11/17 Autocorrelation of Each Frame Let s(k) be a frame of size 128. 1 128 s(k): s(k-h): h=30 x(30) = dot prod. of overlapped = sum(s(31:128).*s(1:99) Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. Autocorrelation x(h): 2018/11/17 Pitch period 30

35 Autocorrelation via DSP Blockset
2018/11/17 Autocorrelation via DSP Blockset Real-time autocorrelation demo: Exercise: Construct the above model and try it. File name pitch1.mdl 2018/11/17

36 Pitch Tracking via Autocorrelation
2018/11/17 Pitch Tracking via Autocorrelation Real-time pitch tracking via autocorrelation: pitch2.mdl Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

37 Formant Analysis Characteristics of formants: Computation methods:
2018/11/17 Formant Analysis Characteristics of formants: Formants are perceptually defined. The corresponding physical property is the frequencies of resonances of the vocal tract. Formant analysis is useful as the position of the first two formants pretty much identifies a vowel. Computation methods: Peak picking on the smoothed spectrum Peak picking on the LP spectrum Factoring for the LP roots Fitting of mixture of Gaussians Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

38 Formant Analysis Track Draw: Formant Location Algorithm
2018/11/17 Formant Analysis Track Draw: A package for formant synthesis with options to sketch formant tracks on a spectrogram. Formant Location Algorithm MATLAB code by Michelle Jamrozik Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

39 Speech Waveform Coding
2018/11/17 Speech Waveform Coding Time domain coding PCM: Pulse Code Modulation DPCM: Differential PCM ADPCM: Adaptive Differential PCM (dspadpcm.mdl) Frequency domain coding Sub-band coding Transform coding Speech Coding in MATLAB Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

40 Conclusions Ideal tools for speech/audio signal processing:
2018/11/17 Conclusions Ideal tools for speech/audio signal processing: MATLAB Simulink Signal Processing Toolbox DSP Blockset Advantages: Reliable functions: well-established and tested Visible graphical algorithm design tools High-level programming language yet C-compatible Powerful visualization capabilities Easy debugging Integrated environment Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17

41 2018/11/17 References [1] “Discrete-Time Processing of Speech Signals”, by Deller, Proakis and Hansen, Prentice Hall, 1993 [2] “Fundamentals of Speech Recognition”, by Rabiner and Juang, Prentice Hall, 1993 [3] “Effects Explained”, [4] “TrackDraw”, [5] “Speech Coding in MATLAB”, Specifically, this is the outline of the talk. Wel start from the basics, introduce the concepts of fuzzy sets and membership functions. By using fuzzy sets, we can formulate fuzzy if-then rules, which are commonly used in our daily expressions. We can use a collection of fuzzy rules to describe a system behavior; this forms the fuzzy inference system, or fuzzy controller if used in control systems. In particular, we can can apply neural networks?learning method in a fuzzy inference system. A fuzzy inference system with learning capability is called ANFIS, stands for adaptive neuro-fuzzy inference system. Actually, ANFIS is already available in the current version of FLT, but it has certain restrictions. We are going to remove some of these restrictions in the next version of FLT. Most of all, we are going to have an on-line ANFIS block for SIMULINK; this block has on-line learning capability and it ideal for on-line adaptive neuro-fuzzy control applications. We will use this block in our demos; one is inverse learning and the other is feedback linearization. 2018/11/17


Download ppt "Speech/Audio Signal Processing in MATLAB/Simulink"

Similar presentations


Ads by Google