Speech Recognition Analysis

1037 Words5 Pages

Abstract—Speech is the most common and natural form of communication among human beings. In the world, there are various languages that are spoken by human being for efficient communication. Then peoples also like to communicate with machine. This can be done by speech recognition. In speech recognition, feature extraction is the most important phase. It is considered as the heart of the system .The work of this is to extract those features from the input speech that help the system in identifying the speech. Main objective of this paper is analysis and summarize mostly widely used feature extraction techniques like Linear predictive coding(LPC).Linear predictive cepstral coefficient (LPCC),Perceptual linear prediction (PLP) and Mel frequency …show more content…

These features can be obtained from spectrogram of the speech signal which is call as short term spectral features. Formants are one of the important features used in speech recognition. Peaks denote dominant frequency components in speech signal. Formants are the peaks of the spectral envelope. The resonance frequency also called formants are inversely proportional to the vocal tract length. These carry the identity of the sound. Pitch is another type of feature. It originates from the vocal cords. When air flow from glottal through the vocal cords, the vibration of the vocal cords produce pitch harmonics. Rate at which the vocal folds vibrate is the frequency of the pitch. So when the vocal folds oscillate at 300 times per second ,they are said to be producing a pitch of 300 Hz .Some other features are voiced and unvoiced information ,short term energy and zero crossing …show more content…

I t is one of the most powerful speech analysis technique. It is one of the useful methods for encoding an analog signal which is human speech is produced in the vocal tract which can be approximated as variable diameter tube.The linear predictive model is based on mathematical approximation of the vocal tract represented by this tube of a varying diameter. At a particular time t , the speech sample S(t) is represented as a linear sum of the P previous samples. The most important aspect of LPC is the linear predictive filter which allows the value of the next sample to be determined by a linear combination of previous

Open Document