- Departments / Centres
Pre-Requisite :EC201 and EC202
Contact Hours and Credits : ( 3 -0- 0 ) 3
The purpose of this course is to explain how DSP techniques could be used for solving problems in speech communication.
Phonetic Representation of speech - Models of Speech production - Perception of Loudness - Critical bands - Pitch perception - Auditory masking.
Short time Energy and Zero-crossing rate - Short time Autocorrelation function - Short Time Fourier transform - The speech spectrogram - Relation of STFT to STACF with speech signals .Shot-Time Cepstrum - Shot time Homo morphic Filtering of Speech signal - Application to pitch detection and Pattern recognition.
Linear prediction and the speech model - Computing the prediction co-efficient-LPC spectrum - Applications to speech compression and pattern recognition.
Digital speech coding - Closed loop coders-Open loop coders - Frequency domain coders. Text to Speech (TTS) analysis - Evolution of speech synthesis systems - Unit selection methods - TTS Applications.
Automatic speech recognition (ASR) - The Decision processes in ASR - Representative recognition performance - Principle Component Analysis- Singular Value Decomposition - Usage of Artificial Intelligence and Linear algebra in Speech processing.
Students are able to