Publications

Refine Results

(Filters Applied) Clear All

Estimating and evaluating confidence for forensic speaker recognition

Published in:
Proc. IEEE Int. Conf. on Acoustics, Speech, and Signal Processing, ICASSP, Vol. 1, 19-23 March 2005, pp. I-717 - I-720.

Summary

Estimating and evaluating confidence has become a key aspect of the speaker recognition problem because of the increased use of this technology in forensic applications. We discuss evaluation measures for speaker recognition and some of their properties. We then propose a framework for confidence estimation based upon scores and metainformation, such as utterance duration, channel type, and SNR. The framework uses regression techniques with multilayer perceptrons to estimate confidence with a data-driven methodology. As an application, we show the use of the framework in a speaker comparison task drawn from the NIST 2000 evaluation. A relative comparison of different types of meta-information is given. We demonstrate that the new framework can give substantial improvements over standard distribution methods of estimating confidence.
READ LESS

Summary

Estimating and evaluating confidence has become a key aspect of the speaker recognition problem because of the increased use of this technology in forensic applications. We discuss evaluation measures for speaker recognition and some of their properties. We then propose a framework for confidence estimation based upon scores and metainformation...

READ MORE

Fusing discriminative and generative methods for speaker recognition: experiments on switchboard and NFI/TNO field data

Published in:
ODYSSEY 2004, Speaker and Language Recognition Workshop, 31 May - 3 June 2004.

Summary

Discriminatively trained support vector machines have recently been introduced as a novel approach to speaker recognition. Support vector machines (SVMs) have a distinctly different modeling strategy in the speaker recognition problem. The standard Gaussian mixture model (GMM) approach focuses on modeling the probability density of the speaker and the background (a generative approach). In contrast, the SVM models the boundary between the classes. Another interesting aspect of the SVM is that it does not directly produce probabilistic scores. This poses a challenge for combining results with a GMM. We therefore propose strategies for fusing the two approaches. We show that the SVM and GMM are complementary technologies. Recent evaluations by NIST (telephone data) and NFI/TNO (forensic data) give a unique opportunity to test the robustness and viability of fusing GMM and SVM methods. We show that fusion produces a system which can have relative error rates 23% lower than individual systems.
READ LESS

Summary

Discriminatively trained support vector machines have recently been introduced as a novel approach to speaker recognition. Support vector machines (SVMs) have a distinctly different modeling strategy in the speaker recognition problem. The standard Gaussian mixture model (GMM) approach focuses on modeling the probability density of the speaker and the background...

READ MORE

The MMSR bilingual and crosschannel corpora for speaker recognition research and evaluation

Summary

We describe efforts to create corpora to support and evaluate systems that meet the challenge of speaker recognition in the face of both channel and language variation. In addition to addressing ongoing evaluation of speaker recognition systems, these corpora are aimed at the bilingual and crosschannel dimensions. We report on specific data collection efforts at the Linguistic Data Consortium, the 2004 speaker recognition evaluation program organized by the National Institute of Standards and Technology (NIST), and the research ongoing at the US Federal Bureau of Investigation and MIT Lincoln Laboratory. We cover the design and requirements, the collections and evaluation integrating discussions of the data preparation, research, technology development and evaluation on a grand scale.
READ LESS

Summary

We describe efforts to create corpora to support and evaluate systems that meet the challenge of speaker recognition in the face of both channel and language variation. In addition to addressing ongoing evaluation of speaker recognition systems, these corpora are aimed at the bilingual and crosschannel dimensions. We report on...

READ MORE

Conversational telephone speech corpus collection for the NIST speaker recognition evaluation 2004

Published in:
Proc. Language Resource Evaluation Conf., LREC, 24-30 May 2004, pp. 587-590.

Summary

This paper discusses some of the factors that should be considered when designing a speech corpus collection to be used for text independent speaker recognition evaluation. The factors include telephone handset type, telephone transmission type, language, and (non-telephone) microphone type. The paper describes the design of the new corpus collection being undertaken by the Linguistic Data Consortium (LDC) to support the 2004 and subsequent NIST speech recognition evaluations. Some preliminary information on the resulting 2004 evaluation test set is offered.
READ LESS

Summary

This paper discusses some of the factors that should be considered when designing a speech corpus collection to be used for text independent speaker recognition evaluation. The factors include telephone handset type, telephone transmission type, language, and (non-telephone) microphone type. The paper describes the design of the new corpus collection...

READ MORE

The mixer corpus of multilingual, multichannel speaker recognition data

Published in:
Proc. Language Resource Evaluation Conf., LREC, 24-30 May 2004, pp. 627-630.

Summary

This paper describes efforts to create corpora to support and evaluate systems that perform speaker recognition where channel and language may vary. Beyond the ongoing evaluation of speaker recognition systems, these corpora are aimed at the bilingual and cross channel dimensions. We report on specific data collection efforts at the Linguistic Data Consortium and the research ongoing at the US Federal Bureau of Investigation and MIT Lincoln Laboratories. We cover the design and requirements, the collections and final properties of the corpus integrating discussions of the data preparation, research, technology development and evaluation on a grand scale.
READ LESS

Summary

This paper describes efforts to create corpora to support and evaluate systems that perform speaker recognition where channel and language may vary. Beyond the ongoing evaluation of speaker recognition systems, these corpora are aimed at the bilingual and cross channel dimensions. We report on specific data collection efforts at the...

READ MORE

High-level speaker verification with support vector machines

Published in:
Proc. IEEE Int. Conf. on Acoustics, Speech and Signal Processing, Vol. 1, ICASSP, 17-21 May 2004, pp. I-73 - I-76.

Summary

Recently, high-level features such as word idiolect, pronunciation, phone usage, prosody, etc., have been successfully used in speaker verification. The benefit of these features was demonstrated in the NIST extended data task for speaker verification; with enough conversational data, a recognition system can become familiar with a speaker and achieve excellent accuracy. Typically, high-level-feature recognition systems produce a sequence of symbols from the acoustic signal and then perform recognition using the frequency and co-occurrence of symbols. We propose the use of support vector machines for performing the speaker verification task from these symbol frequencies. Support vector machines have been applied to text classification problems with much success. A potential difficulty in applying these methods is that standard text classification methods tend to smooth frequencies which could potentially degrade speaker verification. We derive a new kernel based upon standard log likelihood ratio scoring to address limitations of text classification methods. We show that our methods achieve significant gains over standard methods for processing high-level features.
READ LESS

Summary

Recently, high-level features such as word idiolect, pronunciation, phone usage, prosody, etc., have been successfully used in speaker verification. The benefit of these features was demonstrated in the NIST extended data task for speaker verification; with enough conversational data, a recognition system can become familiar with a speaker and achieve...

READ MORE

Multisensor MELPE using parameter substitution

Published in:
Proc. IEEE Int. Conf. on Acoustics, Speech and Signal Processing, ICASSP, Vol. 1, 17-21 May 2004, pp. I-477 - I-480.

Summary

The estimation of speech parameters and the intelligibility of speech transmitted through low-rate coders, such as MELP, are severely degraded when there are high levels of acoustic noise in the speaking environment. The application of nonacoustic and nontraditional sensors, which are less sensitive to acoustic noise than the standard microphone, is being investigated as a means to address this problem. Sensors being investigated include the General Electromagnetic Motion Sensor (GEMS) and the Physiological Microphone (P-mic). As an initial effort in this direction, a multisensor MELPe coder using parameter substitution has been developed, where pitch and voicing parameters are obtained from GEMS and PMic sensors, respectively, and the remaining parameters are obtained as usual from a standard acoustic microphone. This parameter substitution technique is shown to produce significant and promising DRT intelligibility improvements over the standard 2400 bps MELPe coder in several high-noise military environments. Further work is in progress aimed at utilizing the nontraditional sensors for additional intelligibility improvements and for more effective lower rate coding in noise.
READ LESS

Summary

The estimation of speech parameters and the intelligibility of speech transmitted through low-rate coders, such as MELP, are severely degraded when there are high levels of acoustic noise in the speaking environment. The application of nonacoustic and nontraditional sensors, which are less sensitive to acoustic noise than the standard microphone...

READ MORE

Beyond cepstra: exploiting high-level information in speaker recognition

Summary

Traditionally speaker recognition techniques have focused on using short-term, low-level acoustic information such as cepstra features extracted over 20-30 ms windows of speech. But speech is a complex behavior conveying more information about the speaker than merely the sounds that are characteristic of his vocal apparatus. This higher-level information includes speaker-specific prosodics, pronunciations, word usage and conversational style. In this paper, we review some of the techniques to extract and apply these sources of high-level information with results from the NIST 2003 Extended Data Task.
READ LESS

Summary

Traditionally speaker recognition techniques have focused on using short-term, low-level acoustic information such as cepstra features extracted over 20-30 ms windows of speech. But speech is a complex behavior conveying more information about the speaker than merely the sounds that are characteristic of his vocal apparatus. This higher-level information includes...

READ MORE

Exploiting nonacoustic sensors for speech enhancement

Summary

Nonacoustic sensors such as the general electromagnetic motion sensor (GEMS), the physiological microphone (P-mic), and the electroglottograph (EGG) offer multimodal approaches to speech processing and speaker and speech recognition. These sensors provide measurements of functions of the glottal excitation and, more generally, of the vocal tract articulator movements that are relatively immune to acoustic disturbances and can supplement the acoustic speech waveform. This paper describes an approach to speech enhancement that exploits these nonacoustic sensors according to their capability in representing specific speech characteristics in different frequency bands. Frequency-domain sensor phase, as well as magnitude, is found to contribute to signal enhancement. Preliminary testing involves the time-synchronous multi-sensor DARPA Advanced Speech Encoding Pilot Speech Corpus collected in a variety of harsh acoustic noise environments. The enhancement approach is illustrated with examples that indicate its applicability as a pre-processor to low-rate vocoding and speaker authentication, and for enhanced listening from degraded speech.
READ LESS

Summary

Nonacoustic sensors such as the general electromagnetic motion sensor (GEMS), the physiological microphone (P-mic), and the electroglottograph (EGG) offer multimodal approaches to speech processing and speaker and speech recognition. These sensors provide measurements of functions of the glottal excitation and, more generally, of the vocal tract articulator movements that are...

READ MORE

Multimodal speaker authentication using nonacuostic sensors

Published in:
Proc. Workshop on Multimodal User Authentication, 11-12 December 2003, pp. 215-222.

Summary

Many nonacoustic sensors are now available to augment user authentication. Devices such as the GEMS (glottal electromagnetic micro-power sensor), the EGG (electroglottograph), and the P-mic (physiological mic) all have distinct methods of measuring physical processes associated with speech production. A potential exciting aspect of the application of these sensors is that they are less influenced by acoustic noise than a microphone. A drawback of having many sensors available is the need to develop features and classification technologies appropriate to each sensor. We therefore learn feature extraction based on data. State of the art classification with Gaussian Mixture Models and Support Vector Machines is then applied for multimodal authentication. We apply our techniques to two databases--the Lawrence Livermore GEMS corpus and the DARPA Advanced Speech Encoding Pilot corpus. We show the potential of nonacoustic sensors to increase authentication accuracy in realistic situations.
READ LESS

Summary

Many nonacoustic sensors are now available to augment user authentication. Devices such as the GEMS (glottal electromagnetic micro-power sensor), the EGG (electroglottograph), and the P-mic (physiological mic) all have distinct methods of measuring physical processes associated with speech production. A potential exciting aspect of the application of these sensors is...

READ MORE