Publications

Refine Results

(Filters Applied) Clear All

Conditional pronunciation modeling in speaker detection

April 6, 2003
  |
Conference Paper
Author:
David Klusacek
…
Published in:
Proc. IEEE Int. Conf. on Acoustics, Speech and Signal Processing, ICASSP, 6-10 April 2003.
Topic:
speaker recognition
R&D area:
R&D group:

Summary

In this paper, we present a conditional pronunciation modeling method for the speaker detection task that does not rely on acoustic vectors. Aiming at exploiting higher-level information carried by the speech signal, it uses time-aligned streams of phones and phonemes to model a speaker's specific Pronunciation. Our system uses phonemes drawn from a lexicon of pronunciations of words recognized by an automatic speech recognition system to generate the phoneme stream and an open-loop phone recognizer to generate a phone stream. The phoneme and phone streams are aligned at the frame level and conditional probabilities of a phone, given a phoneme, are estimated using co-occurrence counts. A likelihood detector is then applied to these probabilities. Performance is measured using the NIST Extended Data paradigm and the Switchboard-I corpus. Using 8 training conversations for enrollment, a 2.1% equal error rate was achieved. Extensions and alternatives, as well as fusion experiments, are presented and discussed.
READ LESS

Summary

In this paper, we present a conditional pronunciation modeling method for the speaker detection task that does not rely on acoustic vectors. Aiming at exploiting higher-level information carried by the speech signal, it uses time-aligned streams of phones and phonemes to model a speaker's specific Pronunciation. Our system uses phonemes...

READ MORE
Conditional pronunciation modeling in speaker detection

Phonetic speaker recognition using maximum-likelihood binary-decision tree models

April 6, 2003
  |
Conference Paper
Author:
Jiri Navratil
…
Published in:
Proc. IEEE Int. Conf. on Acoustics, Speech, and Signal Processing, ICASSP, Vol. 4, 6-10 April 2003.
Topic:
speaker recognition
R&D area:
R&D group:

Summary

Recent work in phonetic speaker recognition has shown that modeling phone sequences using n-grams is a viable and effective approach to speaker recognition, primarily aiming at capturing speaker-dependent pronunciation and also word usage. This paper describes a method involving binary-tree-structured statistical models for extending the phonetic context beyond that of standard n-grams (particularly bigrams) by exploiting statistical dependencies within a longer sequence window without exponentially increasing the model complexity, as is the case with n-grams. Two ways of dealing with data sparsity are also studied, namely, model adaptation and a recursive bottom-up smoothing of symbol distributions. Results obtained under a variety of experimental conditions using the NIST 2001 Speaker Recognition Extended Data Task indicate consistent improvements in equal-error rate performance as compared to standard bigram models. The described approach confirms the relevance of long phonetic context in phonetic speaker recognition and represents an intermediate stage between short phone context and word-level modeling without the need for any lexical knowledge, which suggests its language independence.
READ LESS

Summary

Recent work in phonetic speaker recognition has shown that modeling phone sequences using n-grams is a viable and effective approach to speaker recognition, primarily aiming at capturing speaker-dependent pronunciation and also word usage. This paper describes a method involving binary-tree-structured statistical models for extending the phonetic context beyond that of...

READ MORE
Phonetic speaker recognition using maximum-likelihood binary-decision tree models

The SuperSID project : exploiting high-level information for high-accuracy speaker recognition

April 6, 2003
  |
Conference Paper
Author:
Douglas A. Reynolds
…
Published in:
Proc. IEEE Int. Conf. on Acoustics, Speech and Signal Processing, ICASSP, Vol. 4, 6-10 April 2003, pp. IV-784 - IV-787.
Topic:
speaker recognition
R&D area:
R&D group:

Summary

The area of automatic speaker recognition has been dominated by systems using only short-term, low-level acoustic information, such as cepstral features. While these systems have indeed produced very low error rates, they ignore other levels of information beyond low-level acoustics that convey speaker information. Recently published work has shown examples that such high-level information can be used successfully in automatic speaker recognition systems and has the potential to improve accuracy and add robustness. For the 2002 JHU CLSP summer workshop, the SuperSID project was undertaken to exploit these high-level information sources and dramatically increase speaker recognition accuracy on a defined NIST evaluation corpus and task. This paper provides an overview of the structures, data, task, tools, and accomplishments of this project. Wide ranging approaches using pronunciation models, prosodic dynamics, pitch and duration features, phone streams, and conversational interactions were explored and developed. In this paper we show how these novel features and classifiers indeed provide complementary information and can be fused together to drive down the equal error rate on the 2001 NIS extended data task to 0.2% - a 71% relative reduction in error over the previous state of the art.
READ LESS

Summary

The area of automatic speaker recognition has been dominated by systems using only short-term, low-level acoustic information, such as cepstral features. While these systems have indeed produced very low error rates, they ignore other levels of information beyond low-level acoustics that convey speaker information. Recently published work has shown examples...

READ MORE
The SuperSID project : exploiting high-level information for high-accuracy speaker recognition

Using prosodic and conversational features for high-performance speaker recognition : report from JHU WS'02

April 6, 2003
  |
Conference Paper
Author:
B. Peskin
…
Published in:
Proc. IEEE Int. Conf. on Acoustics, speech, and Signal Processing, ICASSP, Vol. IV, 6-10 April 2003, pp. IV-792 - IV-795.
Topic:
speaker recognition
R&D area:
R&D group:

Summary

While there has been a long tradition of research seeking to use prosodic features, especially pitch, in speaker recognition systems, results have generally been disappointing when such features are used in isolation and only modest improvements have been set when used in conjunction with traditional cepstral GMM systems. In contrast, we report here on work from the JHU 2002 Summer Workshop exploring a range of prosodic features, using as testbed NIST's 2001 Extended Data task. We examined a variety of modeling techniques, such as n-gram models of turn-level prosodic features and simple vectors of summary statistics per conversation side scored by kth nearest-neighbor classifiers. We found that purely prosodic models were able to achieve equal error rates of under 10%, and yielded significant gains when combined with more traditional systems. We also report on exploratory work on "conversational" features, capturing properties of the interaction across conversion sides, such as turn-taking patterns.
READ LESS

Summary

While there has been a long tradition of research seeking to use prosodic features, especially pitch, in speaker recognition systems, results have generally been disappointing when such features are used in isolation and only modest improvements have been set when used in conjunction with traditional cepstral GMM systems. In contrast...

READ MORE
Using prosodic and conversational features for high-performance speaker recognition : report from JHU WS'02

A multi-threaded fast convolver for dynamically parallel image filtering

March 1, 2003
  |
Journal Article
Author:
Jeremy Kepner
Published in:
J. Parallel Distrib. Comput, Vol. 63, No. 3, March 2003, pp. 360-372.
Topic:
image processing
R&D area:
R&D group:

Summary

2D convolution is a staple of digital image processing. The advent of large format imagers makes it possible to literally ''pave'' with silicon the focal plane of an optical sensor, which results in very large images that can require a significant amount computation to process. Filtering of large images via 2D convolutions is often complicated by a variety of effects (e.g., non-uniformities found in wide field of view instruments) which must be compensated for in the filtering process by changing the filter across the image. This paper describes a fast (FFT based) method for convolving images with slowly varying filters. A parallel version of the method is implemented using a multi-threaded approach, which allows more efficient load balancing and a simpler software architecture. The method has been implemented within a high level interpreted language (IDL), while also exploiting open standards vector libraries (VSIPL) and open standards parallel directives (OpenMP). The parallel approach and software architecture are generally applicable to a variety of algorithms and has the advantage of enabling users to obtain the convenience of an easy operating environment while also delivering high performance using a fully portable code.
READ LESS

Summary

2D convolution is a staple of digital image processing. The advent of large format imagers makes it possible to literally ''pave'' with silicon the focal plane of an optical sensor, which results in very large images that can require a significant amount computation to process. Filtering of large images via...

READ MORE
A multi-threaded fast convolver for dynamically parallel image filtering

Phonetic speaker recognition with support vector machines

January 1, 2003
  |
Conference Paper
Author:
William M. Campbell
…
Published in:
Adv. in Neural Information Processing Systems 16, 2003 Conf., 8-13 December 2003, p. 1377-1384.
Topic:
speaker recognition
R&D area:
R&D group:

Summary

A recent area of significant progress in speaker recognition is the use of high level features-idiolect, phonetic relations, prosody, discourse structure, etc. A speaker not only has a distinctive acoustic sound but uses language in a characteristic manner. Large corpora of speech data available in recent years allow experimentation with long term statistics of phone patterns, word patterns, etc. of an individual. We propose the use of support vector machines and term frequency analysis of phone sequences to model a given speaker. To this end, we explore techniques for text categorization applied to the problem. We derive a new kernel based upon a linearization of likelihood ratio scoring. We introduce a new phone-based SVM speaker recognition approach that halves the error rate of conventional phone-based approaches.
READ LESS

Summary

A recent area of significant progress in speaker recognition is the use of high level features-idiolect, phonetic relations, prosody, discourse structure, etc. A speaker not only has a distinctive acoustic sound but uses language in a characteristic manner. Large corpora of speech data available in recent years allow experimentation with...

READ MORE
Phonetic speaker recognition with support vector machines

Modeling prosodic dynamics for speaker recognition

January 1, 2003
  |
Conference Paper
Author:
A. G. Adami
…
Published in:
Proc. IEEE Int. Conf. on Acoustics, Speech and Signal Processing, ICASSP, Vol. 4, 6-10 April 2003, pp. IV-788 - IV-791.
Topic:
speaker recognition
R&D area:
R&D group:

Summary

Most current state-of-the-art automatic speaker recognition systems extract speaker-dependent features by looking at short-term spectral information. This approach ignores long-term information that can convey supra-segmental information, such as prosodics and speaking style. We propose two approaches that use the fundamental frequency and energy trajectories to capture long-term information. The first approach uses bigram models to model the dynamics of the fundamental frequency and energy trajectories for each speaker. The second approach uses the fundamental frequency trajectories of a pre-defined set of works as the speaker templates and then, using dynamic time warping, computes the distance between templates and the works from the test message. The results presented in this work are on Switchboard 1 using the NIS extended date evaluation design. We show that these approaches can achieve an equal error rate of 3.7% which is a 77% relative improvement over a system based on short-term pitch and energy features alone.
READ LESS

Summary

Most current state-of-the-art automatic speaker recognition systems extract speaker-dependent features by looking at short-term spectral information. This approach ignores long-term information that can convey supra-segmental information, such as prosodics and speaking style. We propose two approaches that use the fundamental frequency and energy trajectories to capture long-term information. The first...

READ MORE
Modeling prosodic dynamics for speaker recognition

Cluster detection in databases : the adaptive matched filter algorithm and implementation

January 1, 2003
  |
Journal Article
Author:
Jeremy Kepner
…
Published in:
Data Mining and Knowledge Discovery, Vol. 7, No. 1, January 2003, pp. 57-79.
Topic:
signal processing
R&D area:
R&D group:

Summary

Matched filter techniques are a staple of modern signal and image processing. They provide a firm foundation (both theoretical and empirical) for detecting and classifying patterns in statistically described backgrounds. Application of these methods to databases has become increasingly common in certain fields (e.g. astronomy). This paper describes an algorithm (based on statistical signal processing methods), a software architecture (based on a hybrid layered approach) and a parallelization scheme (based on a client/server model) for finding clusters in large astronomical databases. The method has proved successful in identifying clusters in real and simulated data. The implementation is flexible and readily executed in parallel on a network of workstations.
READ LESS

Summary

Matched filter techniques are a staple of modern signal and image processing. They provide a firm foundation (both theoretical and empirical) for detecting and classifying patterns in statistically described backgrounds. Application of these methods to databases has become increasingly common in certain fields (e.g. astronomy). This paper describes an algorithm...

READ MORE
Cluster detection in databases : the adaptive matched filter algorithm and implementation

The effect of identifying vulnerabilities and patching software on the utility of network intrusion detection

October 16, 2002
  |
Conference Paper
Author:
Richard P. Lippmann
…
Published in:
Proc. 5th Int. Symp. on Recent Advances in Intrusion Detection, RAID 2002, 16-18 October 2002, pp. 307-326.
Topic:
intrusion detection & monitoring
R&D area:
R&D group:

Summary

Vulnerability scanning and installing software patches for known vulnerabilities greatly affects the utility of network-based intrusion detection systems that use signatures to detect system compromises. A detailed timeline analysis of important remote-to-local vulnerabilities demonstrates (1) vulnerabilities in widely-used server software are discovered infrequently (at most 6 times a year) and (2) Software patches to prevent vulnerabilities from being exploited are available before or simultaneously with signatures. Signature-based intrusion detection systems will thus never detect successful system compromises on small secure sites when patches are installed as soon as they are available. Network intrusion detection systems may detect successful system compromises on large sites where it is impractical to eliminate all known vulnerabilities. On such sites, information from vulnerability scanning can be used to prioritize the large numbers of extraneous alerts caused by failed attacks and normal background MIC. On one class B network with roughly 10 web servers, this approach successfully filtered out 95% of all remote-to-local alerts.
READ LESS

Summary

Vulnerability scanning and installing software patches for known vulnerabilities greatly affects the utility of network-based intrusion detection systems that use signatures to detect system compromises. A detailed timeline analysis of important remote-to-local vulnerabilities demonstrates (1) vulnerabilities in widely-used server software are discovered infrequently (at most 6 times a year) and...

READ MORE
The effect of identifying vulnerabilities and patching software on the utility of network intrusion detection

2-D processing of speech with application to pitch estimation

September 16, 2002
  |
Conference Paper
Author:
Thomas F. Quatieri
Published in:
7th Int. Conf. on Spoken Language Processing, ICSLP 2002, 16-20 September 2002.
Topic:
speech enhancement
R&D area:
R&D group:

Summary

In this paper, we introduce a new approach to two-dimensional (2-D) processing of the one-dimensional (1-D) speech signal in the time-frequency plane. Specifically, we obtain the shortspace 2-D Fourier transform magnitude of a narrowband spectrogram of the signal and show that this 2-D transformation maps harmonically-related signal components to a concentrated entity in the new 2-D plane. We refer to this series of operations as the "grating compression transform" (GCT), consistent with sine-wave grating patterns in the spectrogram reduced to smeared impulses. The GCT forms the basis of a speech pitch estimator that uses the radial distance to the largest peak in the GCT plane. Using an average magnitude difference between pitch-contour estimates, the GCT-based pitch estimator is shown to compare favorably to a sine-wave-based pitch estimator for all-voiced speech in additive white noise. An extension to a basis for two-speaker pitch estimation is also proposed.
READ LESS

Summary

In this paper, we introduce a new approach to two-dimensional (2-D) processing of the one-dimensional (1-D) speech signal in the time-frequency plane. Specifically, we obtain the shortspace 2-D Fourier transform magnitude of a narrowband spectrogram of the signal and show that this 2-D transformation maps harmonically-related signal components to a...

READ MORE
2-D processing of speech with application to pitch estimation
631-640 of 803