John E. Adcock

A closed-form location estimator for use with room environment microphone arrays

The linear intersection (LI) estimator, which is a closed-form method for the localization of source positions given sensor array time-delay estimate information, is presented. The LI estimator is shown to be robust and accurate, to closely model the search-based ML estimator, and to outperform a benchmark algorithm. The computational complexity of the LI estimator is suitable for use in real-time microphone-array applications where search-based location algorithms may be infeasible.

A practical time-delay estimator for localizing speech sources with a microphone array

Abstract A frequency-domain-based delay estimator is described, designed specifically for speech signals in a microphone-array environment. It is shown to be capable of obtaining precision delay estimates over a wide range of signal-to-noise ratio conditions and is computationally simple enough to make it practical for real-time systems. A location algorithm based upon the delay estimator is then developed. With this algorithm it is possible to localize talker positions to a region only a few centimetres in diameter (not very different from the size of the source), and to track a moving source. Experimental results using data from a real 16-element array are presented to indicate the true performance of the algorithms.

A closed-form method for finding source locations from microphone-array time-decay estimates

The linear intersection (LI) estimator, a closed-form method for the localization of source positions given only the sensor array time-delay estimate information, is presented. The array is constrained to be composed of 4-element sub-arrays configured in 2 centered orthogonal pairs. A bearing line in 3-space is estimated from each sub-array and potential source locations are found via closest intersection of bearing line pairs. The final location estimate is determined by a probabilistic weighting of these potential locations. The LI estimator is shown to be robust and accurate, to closely model the ML estimator, and to outperform a representative algorithm. The computational complexity of the LI estimator is suitable for use in real-time microphone-array applications.

A localization-error-based method for microphone-array design

This paper presents a means for predicting the error region associated with a speech-source location estimate obtained from a set of microphones in a room environment. The error predictor presented is derived assuming a specific source-sensor geometry consisting of pairs of closely-spaced sensors for which a delay estimate associated with the potential source has been evaluated. The accuracy of the predictor is evaluated through a set of Monte Carlo simulations and an application of the predictor to microphone-array design in the context of a video-teleconferencing scenario is presented.

Microphone-array speech recognition via incremental map training

For a hidden Markov model (HMM) based speech recognition system it is desirable to combine enhancement of the acoustical signal and statistical representation of model parameters, ensuring both a high quality speech signal and an appropriately trained HMM. In this paper the incremental variant of maximum a posteriori (MAP) estimation is used to adjust the parameters of a talker-independent HMM-based speech recognition system to accurately recognize speech data acquired with a microphone-array. The approach is novel for a microphone-array speech recognition task in that a robust talker-independent model is derived from a baseline system using a relatively small amount of data for training. The results show that (1) ILIAP training significantly improves recognition performance compared to the baseline, and (2) beamforming signal enhancement outperforms single-channel enhancement before and after the adaptive MAP training.

Microphone‐array localization error estimation with application to sensor placement

This paper presents a means for predicting the error region associated with a speech‐source location estimate obtained from a set of microphones in a room environment. The error predictor presented is derived assuming a specific source‐sensor geometry consisting of pairs of closely spaced sensors for which a delay estimate associated with the potential source has been evaluated. The accuracy of the predictor is evaluated through a set of Monte Carlo simulations and an application of the predictor to microphone‐array design in the context of a video‐conferencing scenario is presented.

Practical issues in the use of a frequency‐domain delay estimator for microphone‐array applications

A frequency‐domain delay estimator has been used as the basis of a microphone‐array talker location and beamforming system [M. S. Brandstein and H. F. Silverman, Techn. Rep. LEMS‐116 (1993)]. While the estimator has advantages over previously employed correlation‐based delay estimation methods [H. F. Silverman and S. E. Kirtman, Comput. Speech Lang. 6, 129–152 (1990)], including a shorter analysis window and greater accuracy at lower computational cost, it has the disadvantage that since delays between microphone pairs are estimated independently of one another, there is nothing to ensure that a set of estimated delays corresponds to a single location. This not only introduces errors in talker location but degrades the performance of the beamformer. A method for delay estimation and talker location with a microphone array is described that preserves the low computational complexity and rapid tracking ability of the frequency‐domain delay estimator, while improving the coherence and stability of the estimated delays and derived source locations. Experimental results using data from a real 16‐element array are presented to demonstrate the performance of the algorithms. [Early work principally funded by DARPA/NSF Grant IRI‐8901882, and current work by NSF Grant No. 9314625.]

Utterance dependent parametric warping for a talker-independent HMM-based recognizer

In an effort to improve the recognition performance of talker-independent speech systems, many adaptive methods have been proposed. The methods generally seek to exploit the higher recognition performance rate of talker-dependent systems and extend it to talker-independent systems. This is achieved by some form of placing talkers into several categories, usually using gender or vocal-tract size. We investigate a similar idea, but categorize each utterance independently. An utterance is processed using several spectral compressions, and the compression with the maximum likelihood is then used to train a better model. For testing, the spectral compression with the maximum likelihood is used to decode the utterance. While the spectral compressions divided the utterances well, this did not translate into significant improvement in performance, and the computational cost increase was significant.