Results 1 -
4 of
4
Gammatone Auditory Filterbank and Independent Component Analysis for Speaker Identification systems
, 2005
"... Speaker identification is the process of recognizing who is speaking on the basis of information extracted from the speech signal. It has a number of applications in security and voice controlled service area. However, the most commonly used speaker recognition techniques work successfully only in c ..."
Abstract
- Add to MetaCart
(Show Context)
Speaker identification is the process of recognizing who is speaking on the basis of information extracted from the speech signal. It has a number of applications in security and voice controlled service area. However, the most commonly used speaker recognition techniques work successfully only in clean or matched environment. Accurate speaker identification is made difficult due to a number of factors, with handset/channel mismatch and environmental noise being two of the most prominent. This paper presents a new novel technique which is based on the Gammatone filterbank (GTF) and independent component analysis (ICA). Compared with some other standard techniques, the new technique holds the best performance for a speaker text-independent identification system with mismatched environment and low environmental SNR level (less than 20dB). Key-words: Gammatone filterbank, independent component analysis, speaker identification
International Journal of Innovations in Engineering and Technology (IJIET) Myanmar Continuous Speech Recognition System Based on DTW and HMM
"... Abstract- This paper presents automatic speech recognition for continuous speech in Myanmar Language. Actually, a computer or a machine is not expected to understand what is uttered. But it is expected to be controlled via speech or to transcript the acoustic signal to symbols. This system will also ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract- This paper presents automatic speech recognition for continuous speech in Myanmar Language. Actually, a computer or a machine is not expected to understand what is uttered. But it is expected to be controlled via speech or to transcript the acoustic signal to symbols. This system will also address the issue of automatic word/sentence boundary detection in both quiet and noisy environments. Combinations of LPC, MFCC and GTCC techniques are used in feature extraction. MFCC features give the good discrimination of speech signal. LPC provides an accurate estimate of the speech parameters and it is also an efficient computational model of speech. DTW is used in the feature clustering and HMM is used in the recognition process. The HMM method is extended by combining it with the DTW algorithm in order to combine the advantages of these two powerful pattern recognition technique.
An investigation of non-uniform bandwidths auditory filterbank in audio coding
"... This paper presents an investigation on the use of non-linear auditory filterbank in wideband audio coding. The perceptually based parameterization of the audio signal using gammatone filterbank is examined and discussed. Conventional gammatone filters requires high order FIR filters in the synthesi ..."
Abstract
- Add to MetaCart
(Show Context)
This paper presents an investigation on the use of non-linear auditory filterbank in wideband audio coding. The perceptually based parameterization of the audio signal using gammatone filterbank is examined and discussed. Conventional gammatone filters requires high order FIR filters in the synthesis stage which introduces long delay and large computation cost. Here, a simple and efficient synthesis technique is investigated and embedded into a perceptual audio coder. A limitation of this system is that the outputs of the filterbank can not be maximally decimated due to the non-uniform bandwidths of the filterbank, yet the coder achieves near transparent quality at a bit rate of 120 kbps. 1.