{{{credits}}}
| L | T | P | C |
| 3 | 0 | 0 | 3 |
- To explore the fundamentals of digital speech processing.
- To understand the basic concepts and algorithms of speech processing.
- To familiarize the students with the various speech signal representation, coding and recognition techniques.
- To study the concepts and evaluation method of speech synthesis.
{{{unit}}}
| Unit I | Fundamentals of Digital Speech Processing | 9 |
Introduction: Discrete-Time signals and systems – Transform representation of Signals and systems – Fundamentals of digital filters – Sampling; Process of Speech Production – Acoustic theory of speech production – Digital models for speech signals.
{{{unit}}}
| Unit II | Speech Signal Analysis in Time Domain | 9 |
Time-dependent processing of speech – Methods for extracting the parameters: Energy – Average Magnitude – Zero-crossing rate; Slience discrimation using ZCR and energy – Short-time autocorreleation function – Pitch period estimation using autocorrelation function.
{{{unit}}}
| Unit III | Speech Signal Analysis in Frequency Domain | 9 |
Short time fourier analysis – Fourier transform and linear interpretations – Sampling rates – Spectrographic Displays – Formant extraction – Pitch extraction – Linear predictive coding: Autocorrelation method – Covariance method; Solution of LPC equations – Durbin’s Recursive solution – Application of LPC parameters – Pitch detection.
{{{unit}}}
| Unit IV | Speech Recognition | 9 |
Introduction – Preprocessing – Parametric representation – Speech segmentation – Dynamic time warping – Vector quantization – Hidden Markov Model – Language Models – Developing an isolated digit recognition system.
{{{unit}}}
| Unit V | Speech Synthesis | 9 |
Attributes of speech synthesis – Formant speech synthesis – Concatenative speech synthesis – Prosodic modification of speech – Source filter models for prosody modification – Evaluation of TTS system.
\hfill Total: 45
After the completion of this course, students will be able to:
- Illustrate how the speech production is modelled (K2)
- Extract features from the speech signal in both time and frequency domain (K3)
- Developing a speech recognition system using statistical approach (K3)
- Compare the various methods of speech synthesis (K2)
- L. R. Rabiner and R. W. Schaffer, “Digital Processing of Speech signals”, Prentice Hall, 1978.
- Xuedong Huang, Alex Acero, Hsiao-Wuen Hon, “Spoken Language Processing – A guide to Theory, Algorithm and System Development”, Prentice Hall PTR, 2001.
- Lawrence Rabiner and Biing-Hwang Juang, “Fundamentals of Speech Recognition”, Prentice Hall Signal Processing Series, 1993.
- Thomas F.Quatieri, “Discrete-Time Speech Signal Processing”, Pearson Education, 2002.
- Ben Gold and Nelson Morgan, “Speech and Audio Signal Processing”, John Wiley and Sons Inc., Singapore, 2004.