Hands-free Telephone Research Articles

Performance limits in subband beamforming

This paper analyzes subband beamforming schemes mainly aimed at speech enhancement and acoustic echo suppression applications such as hands-free telephony for both mobile and office environments, Internet telephony and video conferencing. Analytical descriptions of both causal finite-length and noncausal infinite-length subband microphone array structures are given. More specifically, this paper compares finite Wiener filter performance with the noncausal Wiener solution, giving a comprehensive theoretical suppression limit. It is shown that even short filters will yield a good approximation of the infinite solution, provided that the element spacing and temporal sampling is matched to the frequency band of interest. Typically, 10-20 FIR taps are sufficient in each subband.

An exploratory study of language interpretation services provided by videoconferencing.

We explored the feasibility and acceptability of the remote provision of a language interpretation service during general practice consultations. Three methods were used to provide an interpretation service: a physically present interpreter (PI), a remote interpreter accessed using an ISDN videoconferencing link at 128 kbit/s (VI) and a remote interpreter accessed by telephone (TI). Thirty-six non-English-speaking patients were invited to take part and 35 agreed to do so. Twenty-nine (83%) of the patients recruited were female. The age range of the participants was 24-51 years. Fourteen consultations took place with a physically present interpreter, 11 using videoconferencing and 10 using the telephone. Mean scores on the Patient Enablement Instrument were 5.2 for the PI group, 2.3 for the VI group and 5.1 for the TI group. Mean scores on the Medical Interview Satisfaction Scale were 5.3 for the PI group, 4.9 for the VI group and 5.3 for the TI group. The visual quality and sound quality of remote interpreting were satisfactory. Both videoconferencing and hands-free telephones can deliver an acceptable interpreting service in primary care.

Circuit to improve transducer separation in handsfree telephone

In a device having a sound source within an enclosure and a first microphone for picking up sounds outside the enclosure, a system for improving acoustic separation between the sound source and first microphone comprising an additional microphone located approximately equidistant from the sound as the first microphone for picking up sounds within the enclosure, and a circuit connected to the first microphone and the additional microphone for subtracting the sounds within the enclosure from the sounds outside the enclosure.

Background noise cancellation for hands-free communication system of car cabin using adaptive feedforward algorithms

This paper describes the background noise attenuation for hands-free communication systems of car cabins using an adaptive feedforward noise control algorithm. For a hands-free telephone system, the quality of communication usually is affected by background noise such as engine noise and aerodynamic noise. In the present study, an active noise control (ANC) design for reducing car cabin periodic background noise from engine and broadband random noise using adaptive feedforward control is investigated. The control algorithm is based on the well-known filtered-x least mean square (FXLMS) with multi-sine synthesised reference signal and frequency counting technology implemented on a digital signal processor (DSP) platform. Experimental results indicated that the proposed design was effective suppressing 15 dB in periodic engine noise and 2–3 dB in broadband random background noise.

A residual echo cancellation scheme for hands-free telephony

A residual echo cancellation scheme based on the autoregressive (AR) analysis is proposed for hands-free telephony. Residual echo is whitened by the inverse filter using the estimated AR coefficients. Removing the speech characteristics of the residual echo signal, the noise reduction system successfully reduces the power of residual echo as well as that of ambient noise. In a car cabin, the proposed method improved the echo return loss enhancement by 8.7/spl sim/9.9 dB as compared with the conventional combined system.

A psychoacoustic approach to combined acoustic echo cancellation and noise reduction

This paper presents and compares algorithms for combined acoustic echo cancellation and noise reduction for hands-free telephones. A structure is proposed, consisting of a conventional acoustic echo canceler and a frequency domain postfilter in the sending path of the hands-free system. The postfilter applies the spectral weighting technique and attenuates both the background noise and the residual echo which remains after imperfect echo cancellation. Two weighting rules for the postfilter are discussed. The first is a conventional one, known from noise reduction, which is extended to attenuate residual echo as well as noise. The second is a psychoacoustically motivated weighting rule. Both rules are evaluated and compared by instrumental and auditive tests. They succeed about equally well in attenuating the noise and the residual echo. In listening tests, however, the psychoacoustically motivated weighting rule is mostly preferred since it leads to more natural near end speech and to less annoying residual noise.

Integrated echo and noise canceler for hands-free applications

An integrated acoustic echo and noise canceler is proposed for hands-free telephony applications. The proposed system includes a computationally efficient double-talk detection (DTD) method and a residual echo cancellation algorithm. The DTD method uses two cross-correlation coefficients and is robust in adverse conditions. In the residual echo cancellation approach, a linear prediction error filter removes the short-term correlation of the residual echo and the resulting whitened residual echo is canceled by a noise-reduction filter. The proposed system was implemented using a general purpose digital signal processor (DSP) with 16-b arithmetic.

Partitioned residual echo power estimation for frequency‐domain acoustic echo cancellation and postfiltering

AbstractResidual echo arises in hands‐free telephony equipment due to insufficient echo canceler convergence, but can be suppressed using a postfilter. The residual echo power spectral density is the most crucial control parameter for both frequency‐domain acoustic echo cancellation and combined residual echo and noise postfiltering. In this contribution we present and compare residual echo power spectral estimation techniques. We introduce a new partitioned block‐adaptive estimation technique delivering considerably improved residual echo estimates in strongly reverberant and noisy acoustic environments. We show that the adaptation loop of the frequency‐domain adaptive filter (FDAF) can be used simultaneously for residual echo power estimation and tracking of the echo path impulse response. In this way, the FDAF and the postfilter concept supplement each other in a true synergy with low complexity. The resulting echo and noise control system proves to be robust in double talk situations as well.

Optimized second-order gradient microphone for hands-free speech recordings in cars

Hands-free telephony and automatic speech recognition in adverse acoustic conditions require noise reduction and speech enhancement methods to achieve acceptable quality and speech recognition rates. This contribution is primarily focussed on the optimization of the recording microphone. The directivity of the microphone can be used to reduce sound components outside the microphone's main axis which suppresses noise and room reverberation effectively. Considering the outer dimensions, gradient microphones possess a high directivity and good noise suppression properties. In this study, the feasibility and the noise-cancelling characteristics of an optimized second-order gradient microphone are investigated. The development of a simple sensitivity-matching principle allows the use of low-priced microphone capsules. The directivity index, characterizing the diffuse noise suppression, is nearly frequency-independent in the telephone bandwidth and is about 3 dB higher than that of conventional directional microphones used in telecommunication. The microphone is evaluated in an automobile environment with the help of a speaker-dependent single-word recognizer and is compared with lower-order gradient microphones. The noise robustness of the speech recognizer is improved by at least 4.7 dB, provided the microphone is in an optimum position. Conclusions are then drawn regarding such an optimum microphone positioning.

Hands-free telephones – joint control of echo cancellation and postfiltering

This paper deals with an efficient and robust joint control of the step sizes of subband adaptive echo compensation filters and the frequency response of the echo suppression filter of a hands-free telephone system. It is shown that both control methods are described by the same quantity: the ratio of the short-term estimates of the power of the error to the “undisturbed” error signal. Since the latter is not directly measurable, a three-step procedure for its estimation is given. It uses speech activity detection, a “shadow” filter, and a correlation analysis. The procedure described has been implemented. It produced a considerable reduction in the amount of necessary signal processing. Furthermore, it was possible to provide natural communication with hands-free telephone systems.

About time-variant filtering of speech signals with time-frequency distributions for hands-free telephone systems

Joint time-frequency distributions represent the energy or intensity of signal simultaneously in time and frequency. In this paper we introduce a smoothed version of the method denoted as S-method (SM) (IEEE Trans. Signal Process. 42(1) (January 1994) 225–229). It is shown that it is an effective tool for representation of speech signals disturbed by noise. As a consequence, it can be efficiently applied to time-variant filter problems, as they occur in hands-free telephones. The optimal time-variant filter is derived in terms of the smoothed SM, and is illustrated by examples.

Step-size control for acoustic echo cancellation filters – an overview

In this paper we present an overview about several approaches for controlling the step size for adaptive echo cancellation filters in hands-free telephones. First an optimal step size is derived. For the determination of this step size the power of a non-measurable signal has to be estimated. Detection and estimation methods for the determination of this power and for the determination of the optimal step size can be grouped into four classes. For each class, several principles, which differ in their reliability and in their complexity, are presented. Possibilities for combining elements from each class to create an entire step size control unit are also described. An outlook on detector combinations based on neuronal networks concludes this paper.

Adaptation of a memoryless preprocessor for nonlinear acoustic echo cancelling

Acoustic echo cancellers (AECs) in today's hands-free telephones rely on the assumption of a linear echo path. However, low-cost audio equipment or constraints of portable communication systems cause nonlinear distortions in the loudspeaker and its amplifier, which limit the echo reduction of linear AECs. Such an echo path can be modelled by a memoryless nonlinear function preceding the linear FIR filter. Algorithms for joint adaptation of both stages and stepsize normalizations are derived. As examples for the preprocessor a hard-clipping curve and a polynomial are considered. Fast convergence of the latter is achieved with signal orthogonalization or RLS adaptation. Adaptation stepsize control mechanisms are derived using a novel system distance measure. Experiments under adverse conditions and with real hardware demonstrate robust convergence with both models, and an echo reduction improvement by up to 10 dB at amplitude peaks. For a straightforward implementation, computational cost is increased by factor 1.5–4.5 compared to a linear AEC, but ways for complexity reduction are outlined.

GMDF for noise reduction and echo cancellation

A noise-reduction (NR) enhancement to the generalized multi-delay frequency domain algorithm (GMDF) involving minimal additional computations is presented. Findings are presented on the ability of the algorithm to suppress street, office, air-conditioning and car noise from the microphone signal in hands-free telephony. For an additional eleven N/P real multiplies and eight N/P real additions per output sample, where N is the block size and P is the amount of output samples per GMDF iteration, results show an average increase in echo return loss enhancement (ERLE) of 9 dB at low SNR (-5 dB) and 6 dB at higher SNR (15 dB).

State detection for hands-free telephone sets by means of fuzzy LVQ and SOM

The step gain of adaptive algorithms for acoustic echo cancellation should vary with time in dependence of the far-end signal, the local signal, and the adaptation quality. It can be described as state-dependent, which is also the case for the control parameters of other algorithms in a hands-free telephone set, such as adaptive loss control or noise reduction. Most of the known state detection methods can only distinguish between subsets of the states, and are unreliable for certain state transitions, so that they should be combined for improved reliability of the state detection. In this contribution, state detection is performed by a fuzzy learning vector quantization/self-organizing map approach to combine several state detectors. The step gain is inferred from state information using fuzzy logic concepts. Finally, adaptation results are compared to the best conventional step-gain control method used in the complete structure.

Hands-free telephone

The present invention provides a hands-free telephone wherein near-end caller voice detection information is provided with little error, and wherein electric power saving effects are improved, and also wherein the far-end caller does not receive an unnatural impression. An echo canceller comprises: level calculating unit 1, 2, and 3 for calculating the level of reception signals, input transmission signals, and residual signals; and adaptive filter 4 for inputting the reception signals and transmission input signals and outputting residual signals; echo cancellation amount calculation unit 5 for calculating the echo cancellation amount from the input transmission signal level and residual signal level; voice detection threshold value control unit 6 for controlling the voice detection threshold value according to the reception signal level and the echo cancellation amount; voice detection unit 7 for outputting a voice detection flag using the residual signal level and the voice detection threshold value; attenuator control unit 9 for controlling the suppression amount based on the reception signal level, and attenuator 10 for performing suppression according to the suppression amount.

Invisible acoustic screen for open-plan offices and the like

The invisible acoustic screen for reducing sound leakage from handsfree telephones, loudspeaking computer terminals, and the like in a workspace includes a sound reducing apparatus comprising: a sound producing loudspeaker; a cancelling loudspeaker; a sensing microphone having an output; a delay circuit; and, an adaptive filter circuit having a value determined by a transfer function between the sound producing loudspeaker and the sensing microphone and by a transfer function between the cancelling loudspeaker and the sensing microphone; and the output of the sensing microphone being fed back to the adaptive filter circuit; and the received signal coupled through a delay circuit to the sound producing loudspeaker.

Analysis of partitioned frequency-domain LMS adaptive algorithm with application to a hands-free telephone system echo canceller

For adaptive filters with long impulse responses and requiring modelling by an FIR filter, algorithms incorporating computationally efficient DFT-based adaptive block filters are the design of choice. The reduction in computational complexity is very significant in applications such as active noise control and the hands-free telephone system echo cancellation problem where impulse responses of more than a 1000 samples are common. The performance of frequency-domain adaptive algorithms is analysed and, when properly designed, shown to be equivalent to time-domain algorithms with uncorrelated input signals. The design parameters include: block length, DFT length and partitioning of the impulse response. The study includes both constrained and unconstrained parameter adaptation. Guidelines to the design of partitioned frequency-domain LMS (PFLMS) algorithms are given. Copyright © 2000 John Wiley & Sons, Ltd.

Time-varying filtering of speech signals using linear prediction

Time-varying filtering of noisy speech signals based on linear prediction (LP) is presented. The approach is tested on noisy speech signals apparent in hands-free telephone systems.

Acoustic echo control. An application of very-high-order adaptive filters

We have discussed the application of high-order adaptive filters to the problem of acoustical echo cancellation with particular application to hands free telephone systems. We described a means to achieve robust performance. We further presented methods for reducing computational complexity that allow implementation in low-cost, fixed-point digital signal processors. Progress in technology will allow the use of more sophisticated algorithms at lower cost in the near future.