This paper introduces and motivates the use of Gaussian mixture models (GMM) for robust text-independent speaker identification. The individual Gaussian components of a GMM are shown to represent some general speaker-dependent spectral shapes that are effective for modeling speaker identity. The focus of this work is on applications which require high identification rates using short utterance from unconstrained conversational speech and robustness to degradations produced by transmission over a telephone channel. A complete experimental evaluation of the Gaussian mixture speaker model is conducted on a 49 speaker, conversational telephone speech database. The experiments examine algorithmic issues (initialization, variance limiting, model order selection), spectral variability robustness techniques, large population performance, and comparisons to other speaker modeling techniques (uni-modal Gaussian, VQ codebook, tied Gaussian mixture, and radial basis functions). The Gaussian mixture speaker model attains 96.8% identification accuracy using 5 second clean speech utterances and 80.8% accuracy using 15 second telephone speech utterances with a 49 speaker population and is shown to outperform the other speaker modeling techniques on an identical 16 speaker telephone speech task. >

Robust text-independent speaker identification using Gaussian mixture speaker models

https://apps.dtic.mil/sti/pdfs/ADA157550.pdf

The TRACE model of speech perception.

The Hearsay-II system, developed during the DARPA-sponsored five-year speech-understanding research program, represents both a specific solution to the speech-understanding problem and a general framework for coordinating independent processes to achieve cooperative problem-solving behavior. As a computational problem, speech understanding reflects a large number of intrinsically interesting issues. Spoken sounds are achieved by a long chain of successive transformations, from intentions, through semantic and syntactic structuring, to the eventually resulting audible acoustic waves. As a consequence, interpreting speech means effectively inverting these transformations to recover the speaker's intention from the sound. At each step in the interpretive process, ambiguity and uncertainty arise.

The Hearsay-II problem-solving framework reconstructs an intention from hypothetical interpretations formulated at various levels of abstraction. In addition, it allocates limited processing resources first to the most promising incremental actions. The final configuration of the Hearsay-II system comprises problem-solving components to generate and evaluate speech hypotheses, and a focus-of-control mechanism to identify potential actions of greatest value. Many of these specific procedures reveal novel approaches to speech problems. Most important, the system successfully integrates and coordinates all of these independent activities to resolve uncertainty and control combinatorics. Several adaptations of the Hearsay-II framework have already been undertaken in other problem domains, and it is anticipated that this trend will continue; many future systems necessarily will integrate diverse sources of knowledge to solve complex problems cooperatively.

Discussed in this paper are the characteristics of the speech problem in particular, the special kinds of problem-solving uncertainty in that domain, the structure of the Hearsay-II system developed to cope with that uncertainty, and the relationship between Hearsay-II's structure and those of other speech-understanding systems. The paper is intended for the general computer science audience and presupposes no speech or artificial intelligence background.

/pdf/the-hearsay-ii-speech-understanding-system-integrating-4upd3ocgip.pdf

The Hearsay-II Speech-Understanding System: Integrating Knowledge to Resolve Uncertainty

Speech perception: a model of acoustic–phonetic analysis and lexical access

We describe current approaches to text-independent speaker identification based on probabilistic modeling techniques. The probabilistic approaches have largely supplanted methods based on comparisons of long-term feature averages. The probabilistic approaches have an important and basic dichotomy into nonparametric and parametric probability models. Nonparametric models have the advantage of being potentially more accurate models (though possibly more fragile) while parametric models that offer computational efficiencies and the ability to characterize the effects of the environment by the effects on the parameters. A robust speaker-identification system is presented that was able to deal with various forms of anomalies that are localized in time, such as spurious noise events and crosstalk. It is based on a segmental approach in which normalized segment scores formed the basic input for a variety of robust 43% procedures. Experimental results are presented, illustrating 59% the advantages and disadvantages of the different procedures. 64%. We show the role that cross-validation can play in determining how to weight the different sources of information when combining them into a single score. Finally we explore a Bayesian approach to measuring confidence in the decisions made, which enabled us to reject the consideration of certain tests in order to achieve an improved, predicted performance level on the tests that were retained. >

Text-independent speaker identification

HWIM, the speech understanding system developed at BBN as part of the recent five-year ARPA Speech Understanding Research Project, is designed to "understand" naturally spoken utterances relevant to a task domain of travel budget management. Its vocabulary is over 1000 words, and its grammar permits a relatively habitable subset of natural English. HWIM contains sources of knowledge at the levels of acoustic-phonetics, phonology, vocabulary, syntax, semantics, factual knowledge, and discourse. This paper describes the system as a whole and presents its performance results at the end of the project.

The HWIM speech understanding system

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.

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

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.

Investigation of text-independent speaker indentification over telephone channels

In this paper, we present the preliminary performance of four methods for text-independent speaker identification using speech transmitted over radio channels. In a previous paper [1], we showed that for both laboratory-quality and simulated noisy-channel data in a single-session paradigm, new probabilistic classifiers yielded performance superior to that of a minimum distance classifier. We have recently compiled a speech database consisting of speech transmissions over a radio-channel. The lower quality and higher variability of this database differ markedly from the laboratory-quality databases often used in speech processing research. We present preliminary results with the same four methods of text-independent speaker identification using the radio-channel database with several experimental paradigms including multi-session paradigms. These results show that the probabilistic methods perform significantly better than a minimum-distance classifier for the multi-session paradigm.

Further investigation of probabilistic methods for text-independent speaker identification

In this paper, we examine several methods for text-independent speaker identification and discuss the problem of achieving robust performance. In a previous paper [1], we have described a radio-channel database that contains highly variable speech data of poor quality. Here, we describe several experiments with the radio-channel database leading to the development of robust features and methods. These experiments show that robustness to one degradation may not be sufficient to improve speaker identification accuracy. A robust feature set and modeling and classification method should mitigate the effects of many of the degradations in the data.

J. Wolf

Papers

The HWIM speech understanding system

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

Investigation of text-independent speaker indentification over telephone channels

Further investigation of probabilistic methods for text-independent speaker identification

Investigation of text-independent speaker identification techniques under conditions of variable data