M. Krasner

MIT Lincoln Laboratory

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Methods and experiments for text-independent speaker recognition over telephone channels

H. Gish M. Krasner W. Russell Jared J. Wolf

We consider methods for text-independent speaker identification that deal with the variability in the data introduced by unknown telephone channels. The methods investigated include probabilistic channel modeling, a channel-invariant model and a modified-Gaussian model. The methods are described and then evaluated with experiments conducted with a twenty speaker database of long distance telephone calls.

Speaker identification

Identification

10.1109/icassp.1986.1168974

Cite

Citations (30)

Investigation of text-independent speaker indentification over telephone channels

H. Gish K. Karnofsky M. Krasner S. Roucos Richard Schwartz

In this paper, we examine several methods for text-independent speaker identification of telephone speech with limited duration data, The issue addressed is the assessment of channel characteristics, especially linear aspects, and methods for improving speaker identification performance when the speaker to be identified is on a different telephone channel than that data used for training. We show experimental evidence illustrating the cross-channel problem and also show that the direct approach, of using simple channel-invariant features, can discard much speaker dependent information. The methods we have found to be most effective rely on the training process to incorporate channel variability.

Identification

Speaker identification

Speaker diarisation

Training set

10.1109/icassp.1985.1168410

Cite

Citations (29)

Context-dependent modeling for acoustic-phonetic recognition of continuous speech

Richard Schwartz Y.-L. Chow Owen Kimball S. Roucos M. Krasner

This paper describes the results of our work in designing a system for phonetic recognition of unrestricted continuous speech. We describe several algorithms used to recognize phonemes using context-dependent Hidden Markov Models of the phonemes. We present results for several variations of the parameters of the algorithms. In addition, we propose a technique that makes it possible to integrate traditional acoustic-phonetic features into a hidden Markov process. The categorical decisions usually associated with heuristic acoustic-phonetic algorithms are replaced by automated training techniques and global search strategies. The combination of general spectral information and specific acoustic-phonetic features is shown to result in more accurate phonetic recognition than either representation by itself.

Categorical variable

Representation

10.1109/icassp.1985.1168283

Cite

Citations (215)

Time and frequency domain noise shaping in speech coding

J. Makhoul M. Berouti M. Krasner

We present a general framework for the coding of signals with a time-varying short-term spectrum. The total quantization distortion is minimized by a dynamic bit allocation in time and in frequency. The resulting distortion is then white noise with fixed power. Such noise, however, is not optimal perceptually when coding speech signals. Spectral and temporal noise shaping are then used to minimize the perception of noise in frequency and in time. Initial experiments at 16 kb/s demonstrate the validity of shaping the noise in time as well as in frequency.

Colors of noise

Noise power

10.1109/icassp.1981.1171223

Cite

Citations (0)

Grammatical constraints and recognition performance

The Journal of the Acoustical Society of America (1986)

P. Price Y.-L. Chow M. Dunham Owen Kimball M. Krasner

The integration of grammatical with acoustical knowledge sources in the BBN continuous speech recognition system, BYBLOS, and the resulting effects on performance are described. The system consists of feature extraction, acoustical scoring, and linguistic scoring. Feature extraction is based on vector quantized reel-warped cepstral coefficients. Acoustical scoring is derived from a hidden Markov model for each word, where word models are based on phonetic spellings so that models can be computed for words that have never been trained. The linguistic model is represented as a finite automaton derived automatically from a context-free specification of the task-domain syntax and semantics. It is shown how recognition performance varies with properties of the grammars. Word recognition accuracies of over 98% have been achieved in continuous speaker-dependent mode for 350-word tasks with grammars having maximum perplexity in the range of 20 to 60. [Work supported by DARPA and monitored by NAVELEX.]

Perplexity

Feature (linguistics)

10.1121/1.2023686

Cite

Citations (0)

Research on Narrowband Communications

John Makhoul M. Krasner Salim Roukos Richard Evan Schwartz John Sørensen

Abstract : We report on research toward a very-low-rate vocoder. This quarter we continued investigation in three areas. The first area of research is multi- speaker synthesis: speech synthesis from the transmitted vocoder parameters with the voice quality of the vocoder user. This processing entails speaker-specific spectral transformation of the vocoder diphone database. The second area of research is to improve the accuracy of the phonetic recognition. Our work this quarter concentrated on training the recognizer by augmentation of the diphone database with diphones extracted from natural, continuous speech. The third area of research is the development of an efficient model of continuous speech. We have developed a novel method of a variable-order Markov chain. We are continuing evaluation of this method.

Narrowband

10.21236/ada102418

Cite

Citations (0)

to result in more accurate phonetic

S. Roucos M. Krasner

Source

Cite

Citations (0)

Stability analysis of APC systems

M. Krasner M. Berouti J. Makhoul

Adaptive predictive coding (APC) is a useful technique for high-quality digital encoding of speech signals at medium band data rates. A particularly annoying degradation in APC systems, however, is the presence of "glitches" or "beeps" in the output speech. These correspond to frames with signal-to-quantization-noise ratios (S/Q) that are often less than unity, i.e., negative in dB. We show that these degradations are caused by instabilities of the APC system. This paper provides a method for analyzing the stability of APC systems as a function of the encoding bit rate, the quantization algorithm, and the power gain of the APC loop feedback filter. Two methods of improving system performance based on the results of the stability analysis are then developed. Experimental evaluation by listening tests confirm the validity of the techniques.

10.1109/icassp.1981.1171230

Cite

Citations (4)

Continuous Speech Recognition on a Butterfly Parallel Processor.

Proceedings of the International Conference on Parallel Processing (1986)

L. Cosell Owen Kimball Richard Schwartz M. Krasner

Source

Cite

Citations (1)

Research in Continuous Speech Recognition

R. Schwartz Y.-L. Chow Owen Kimball M. Krasner J. Makhoul

10.21236/ada150474

Cite

Citations (0)