DCT assisted speaker identification in the presence of noise and channel degradation

Abstract

This paper presents a robust speaker identification method from degraded speech signals. This proposed method depends on the Mel-frequency cepstral coefficients (MFCCs) for feature extraction from the degraded speech and its discrete cosine transform (DCT). It is known that the MFCCs based speech recognition methods are not robust enough in the presence of noise and channel degradation. So, the feature extraction from the DCT of the signal will assist in achieving a higher recognition rate. The artificial neural network (ANN) classification technique is used in the proposed method. The comparison between the proposed method and the method using the MFCCs only for feature extraction from noisy speech signals and telephone-like degraded signals shows that the proposed method improves the recognition rate in the presence of noise or degradation. , Egypt since 1990. He has published several scientific papers in national and international conferences and journals. His current research areas of interest include image processing, speech processing, digital communications and electromagnetic applications. since 1987. He has published several scientific papers in national and international conferences and journals. His current research areas of interest include adaptive signal processing techniques, image processing, speech processing and digital communications.

An efficient Automatic Speaker Identification System based Pitch Frequency Estimation in Degraded Environmental Conditions

“…Recently, a lot of research works adopted DTs such as DWT [5], DCT [8], and DST [9] for feature extraction. These DTs have an excellent energy compaction property, which is suitable for the elimination of the noise effect.…”

“…These DTs have an excellent energy compaction property, which is suitable for the elimination of the noise effect. Shafik et al [8] proved that the identification of speakers with features extracted from DTs, such as DWT and DCT, gives better recognition rates. Li et al [9] suggested the utilization of the DST in the speaker identification systems.…”

“…The DCT is a trigonometric transform that can be estimated for a speech signal x (n) as follows [8]:…”

Automatic Speaker Identification System based MLP Network and Deep Learning in the Presence of Severe Interference

“…The drawback of the MFCCs is the poor performance with noise. Recently, a lot of research adopted discrete transforms such as DWT [37], DCT [36], and DST [24] for feature extraction. These discrete transforms have an excellent energy compaction property, which is suitable for the elimination of the noise effect.…”

“…These discrete transforms have an excellent energy compaction property, which is suitable for the elimination of the noise effect. Shafik et al [36] and Abd El-Samie et al [3] proved that the identification of speakers with features extracted in transform domains such as DWT, DCT gives better recognition rates. Li et al [24] proposed the utilization of the DST in the speaker identification systems.…”