Cepstrum

About: Cepstrum is a research topic. Over the lifetime, 3346 publications have been published within this topic receiving 55742 citations.

Speaker-independent isolated word recognition based on emphasized spectral dynamics

Abstract: A new speech analysis technique applicable to speech recognition is proposed considering the auditory mechanism of speech perception which emphasizes spectral dynamics and which compensates for the spectral undershoot associated with coarticulation. A speech wave is represented by the LPC cepstrum and logarithmic energy sequences, and the time sequences over short periods are expanded by the first- and second-order polynomial functions at every frame period. The dynamics of the cepstrum sequences are then emphasized by the linear combination of their polynomial expansion coefficients, that is, derivatives, and their instantaneous values. Speaker-independent word recognition experiments using time functions of the dynamics-emphasized cepstrum and the polynomial coefficient for energy indicate that the error rate can be largely reduced by this method.

Product of power spectrum and group delay function for speech recognition

Abstract: Mel-frequency cepstral coefficients (MFCCs) are the most widely used features for speech recognition. These are derived from the power spectrum of the speech signal. Recently, the cepstral features derived from the modified group delay function (MGDF) have been studied by Murthy and Gadde (Proc. ICASSP, vol.1, p.68-71, 2003) for speech recognition. In this paper, we propose to use the product of the power spectrum and the group delay function (GDF), and derive the MFCCs from the product spectrum. This spectrum combines the information from the magnitude spectrum as well as the phase spectrum. The MFCCs of the MGDF are also investigated in this paper. Results show that the cepstral features derived from the power spectrum perform better than that from the MGDF, and the product spectrum based features provide the best performance.

Robust text-independent speaker identification over telephone channels

Abstract: This paper addresses the issue of closed-set text-independent speaker identification from samples of speech recorded over the telephone. It focuses on the effects of acoustic mismatches between training and testing data, and concentrates on two approaches: (1) extracting features that are robust against channel variations and (2) transforming the speaker models to compensate for channel effects. First, an experimental study shows that optimizing the front end processing of the speech signal can significantly improve speaker recognition performance. A new filterbank design is introduced to improve the robustness of the speech spectrum computation in the front-end unit. Next, a new feature based on spectral slopes is described. Its ability to discriminate between speakers is shown to be superior to that of the traditional cepstrum. This feature can be used alone or combined with the cepstrum. The second part of the paper presents two model transformation methods that further reduce channel effects. These methods make use of a locally collected stereo database to estimate a speaker-independent variance transformation for each speech feature used by the classifier. The transformations constructed on this stereo database can then be applied to speaker models derived from other databases. Combined, the methods developed in this paper resulted in a 38% relative improvement on the closed-set 30-s training 5-s testing condition of the NIST'95 Evaluation task, after cepstral mean removal.

Intelligibility of speech with filtered time trajectories of spectral envelopes

Abstract: The effect of filtering the time trajectories of spectral envelopes on speech intelligibility was investigated. Since the LPC cepstrum forms the basis of many automatic speech recognition systems, the authors filtered time trajectories of the LPC cepstrum of speech sounds, and the modified speech was reconstructed after the filtering. For processing, they applied low-pass, high-pass and band-pass filters. The accuracy results from the perceptual experiments for Japanese syllables show that speech intelligibility is not severely impaired as long as the filtered spectral components have 1) a rate of change faster than 1 Hz when high-pass filtered, 2) a rate of change slower than 24 Hz when low-pass filtered, and 3) a rate of change between 1 and 16 Hz when band-pass filtered.

Bangla Speech Recognition System Using LPC and ANN

Abstract: This paper presents the Bangla speech recognition system. Bangla speech recognition system is divided mainly into two major parts. The first part is speech signal processing and the second part is speech pattern recognition technique. The speech processing stage consists of speech starting and end point detection, windowing, filtering, calculating the Linear Predictive Coding(LPC) and Cepstral Coefficients and finally constructing the codebook by vector quantization. The second part consists of pattern recognition system using Artificial Neural Network(ANN). Speech signals are recorded using an audio wave recorder in the normal room environment. The recorded speech signal is passed through the speech starting and end-point detection algorithm to detect the presence of the speech signal and remove the silence and pauses portions of the signals. The resulting signal is then filtered for the removal of unwanted background noise from the speech signals. The filtered signal is then windowed ensuring half frame overlap. After windowing, the speech signal is then subjected to calculate the LPC coefficient and Cepstral coefficient. The feature extractor uses a standard LPC Cepstrum coder, which converts the incoming speech signal into LPC Cepstrum feature space. The Self Organizing Map(SOM) Neural Network makes each variable length LPC trajectory of an isolated word into a fixed length LPC trajectory and thereby making the fixed length feature vector, to be fed into to the recognizer. The structures of the neural network is designed with Multi Layer Perceptron approach and tested with 3, 4, 5 hidden layers using the Transfer functions of Tanh Sigmoid for the Bangla speech recognition system. Comparison among different structures of Neural Networks conducted here for a better understanding of the problem and its possible solutions.