This paper proposes an efficient method for estimating frame energy of speech from EnhancedVariable Rate Coder (EVRC) bitstream for network-based speech processing applications in Transcoder Free Operation (TrFO) environments, where speech signals are represented as speech coding parameters. A frame of speech energy is decomposed into the energy of excitation and vocal tract filter, and the frame energy estimation method is derived for each component. Among many parameters of EVRC bitstream, the fixed codebook gain and adaptive codebook gain are used for the estimation of excitation energy, and Line Spectrum Pair (LSP) information is used to estimate the energy of vocal tract filter. Experimental results demonstrated the novelty of the proposed method. The correlation coefficient between the actual and estimated frame energy can be maintained at a value of 0.994 with just 5% multiplicative operations of full decoding.

In this paper, we propose a novel model for estimating the quality of speech without the reference speech information. The proposed auditory non-intrusive quality estimation plus (ANIQUE+) model is a perceptual model simulating the functional role of human auditory system, and employs improved modeling of quality estimation by statistical learning methods. Experimental evaluation demonstrated that the performance of the ANIQUE+ model is significantly superior to that of the current ITU-T standard recommendation P.563 on 34 different subjective mean opinion score (MOS) databases - the averaged correlation between subjective and objective quality scores is about 0.97 for ANIQUE+, whereas P.563 shows 0.87 averaged correlation.

In the past few years, objective quality assessment models have become increasingly used for assessing or monitoring speech and audio quality. By measuring perceived quality on an easily-understood subjective scale, such as listening quality (excellent, good, fair, poor, bad), these methods provide a quick and repeatable way to estimate customer experience. Typical applications include audio quality evaluation, selection of codecs or other equipment, and measuring the quality of telephone networks. To introduce this special issue, this paper provides an overview of the field, outlining the main approaches to intrusive, nonintrusive and parametric models and discussing some of their limitations and areas of future work

In predicting subjective quality of speech signal degraded by telecommunication networks, conventional objective models require a reference source speech signal, which is applied as an input to the network, as well as the degraded speech. Non-intrusive estimation of speech quality is a challenging problem in that only the degraded speech signal is available. Non-intrusive estimation can be used in many real applications when source speech signal is not available. In this paper, we propose a new approach for non-intrusive speech quality estimation utilizing the temporal envelope representation of speech. The proposed auditory non-intrusive quality estimation (ANIQUE) model is based on the functional roles of human auditory systems and the characteristics of human articulation systems. Experimental evaluations on 35 different tests demonstrated the effectiveness of the proposed model.

In order to estimate the quality of degraded speech processed by communication networks, conventional objective speech quality assessment methods require the source speech signal, which has been applied to the networks, as well as the processed speech. The paper presents a new paradigm in objective speech quality assessment. In contrast to previous objective models, the proposed auditory non-intrusive quality estimation (ANIQUE) model estimates the quality of speech without using the source speech information. ANIQUE is a perceptual model which simulates the functional role of the human auditory system at both peripheral level and central auditory level. The performance of ANIQUE is demonstrated using 19 different subjective MOS databases.

In this paper, we investigate a new paradigm of objective speech quality estimation. The proposed nonintrusive method utilizes only processed speech signal, whereas conventional objective models require source speech applied as an input to the system under test, as well as the processed speech. The proposed method is based on the temporal envelope representation of speech, which reflects the perceptual characteristics of human auditory systems and human speech production systems, and we found it provides a useful cue for nonintrusive objective speech quality estimation. The performance of the proposed method is demonstrated for four different subjective mean opinion score databases.

It is essential to incorporate perceptual characteristics of human hearing in modern speech/audio coding systems. However, the focus has been confined only to the magnitude information of speech, and little attention has been paid to phase information. A quantitative study on the characteristics of human phase perception is presented and a novel method is proposed for the quantization of phase information in speech/audio signals. First, the just-noticeable difference (JND) of phase for each harmonic in flat-spectrum periodic tones is measured for several different fundamental frequencies. Then, a mathematical model of JND is established, based on measured data, to form a weighting function for phase quantization. Since the proposed weighting function is derived from psychoacoustic measurements, it provides a novel quantization method by which more bits are assigned to perceptually important phase components at the sacrifice of less important ones, resulting in a quantized signal perceptually closer to the original one. Experimental results on five vowel speech signals demonstrate that the proposed weighting function is very effective for the quantization of phase information.

We propose a novel speech enhancement algorithm, termed improved global soft decision (IGSD). IGSD is a unified framework for global soft decision on speech absence/presence, noise spectrum estimation, spectral gain modification based on Ephraim-Malah noise suppression. In IGSD, speech absence probability (SAP) is the most important factor, and we propose an efficient and novel SAP estimation in which the SAP is derived based on the general hypothesis for speech absence/presence. In IGSD, the global SAP based on the global hypothesis for speech absence/presence is used to prevent from the problem caused by insufficient amount of data, but more general hypothesis is utilized in the derivation of global SAP estimation. The performance of IGSD is evaluated both subjectively and objectively, and the quality of speech is improved significantly, compared with conventional GSD speech enhancement algorithm.

For efficient quantization of speech representations, it is essential to incorporate perceptual characteristics of human hearing. However, the focus has been confined only to the magnitude information of speech, and little attention has been paid to phase information. This paper presents a novel approach, termed perceptually irrelevant phase elimination (PIPE), to find out irrelevant phase information in acoustic signals in terms of perceived quality. The proposed method, inspired by the observation that the relative phase relationship within a critical band is perceptually important, is derived not only for stationary Fourier signals but also for harmonic signals. For harmonic signals, the "critical phase frequency" is defined below which phase information is perceptually irrelevant. The PIPE algorithm is incorporated into the harmonic analysis/synthesis of speech, and subjective test results demonstrate the effectiveness of the proposed method.

For efficient quantization of speech representation, it is common to incorporate perceptual characteristics of human hearing. However, the focus has been confined only to the magnitude information of speech, and little attention has been paid to phase information. This paper presents a novel approach, termed perceptually irrelevant phase elimination (PIPE), to find out irrelevant phase information of acoustic signals in terms of perception. The proposed method, which is based on the observation that the relative phase relationship within a critical band is perceptually important, is derived not only for stationary Fourier signal but also for harmonic signal. The proposed method is then incorporated into the analysis-by-synthesis system based on harmonic representation of speech, and subjective test results demonstrate its effectiveness.

This paper presents a new approach to an auditory model for robust speech recognition in noisy environments. The proposed model consists of cochlear bandpass filters and nonlinear operations in which frequency information of the signal is obtained by zero-crossing intervals. Intensity information is also incorporated by a peak detector and a compressive nonlinearity. The robustness of the zero-crossings in spectral estimation is verified by analyzing the variance of the level-crossing intervals as a function of the crossing level values. Compared with other auditory models, the proposed auditory model is computationally efficient, free from many unknown parameters, and able to serve as a robust front-end for speech recognition in noisy environments. Experimental results of speech recognition demonstrate the robustness of the proposed method in various types of noisy environments.

SPEECH RECOGNITION Doh-Suk Kimy, Jae-Hoon Jeongy, Soo-Young Leey, Rhee M. Kilz yDepartment of Electrical Engineering/ zDivision of Basic Science Korea Advanced Institute of Science and Technology 373-1 Kusong-dong, Yusong-gu, Taejon 305-701, Korea E-mail: [email protected] ABSTRACT Zero-crossings with peak amplitudes (ZCPA) model motivated by human auditory periphery is simple compared with other auditory models, but powerful speech analysis tool for robust speech recognition in noisy environments. In this paper, improvement in recognition rate of ZCPA model is addressed by incorporating time-derivative features with several di erent time-derivative window lengths. Experimental results show that ZCPA has relatively higher sensitivity to derivative window length than conventional feature extraction algorithms. Also, experimental comparisons with several front-ends including some auditorylike schemes in real-world noisy environments demonstrate the robustness of ZCPA model. ZCPA model shows superior performance compared with other frontends especially in noisy condition corrupted by white Gaussian noise.

The ensemble interval histogram (EIH) is an auditory model which can be used as a robust "front-end" for speech recognition systems. The utilization of multiple level-crossing detectors in the EIH provides frequency and intensity information, which may be useful for speech processing. Proper determination of the number of levels and the level values is very important for reliable performance of the system. An analytic relationship is developed for the variance and SNR of the level-crossing intervals as a function of the crossing level value, and a new feature extraction method based on zero-crossings with peak amplitudes is proposed for robust speech recognition in noisy environments. The proposed method not only can preserve intensity information, but also is robust to noise in estimating the frequency information without the need to determine the level values and the number of levels. Experimental results show the robustness of the proposed method.