Corpus ID: 14891265

A na ve de-lambing method for speaker identification

  title={A na ve de-lambing method for speaker identification},
  author={Qin Jin and A. Waibel},
This paper addresses the issue of close-set text-independent speaker identification from speech samples recorded over telephone. We have known that the speaker identification performance variability can be attributed to many factors. One major factor is the inherent differences in the recognizability of different speakers. In speaker recognition systems such differences are characterized by the use of animal names for different types of speakers. In this paper we use lambs to refer to those… Expand
Finding Difficult Speakers in Automatic Speaker Recognition
The phenomenon that some speakers within a given population have a tendency to cause a large proportion of errors, and ways of finding such speakers are investigated, as well as a straightforward approach to predict speakers that will be difficult for a system to correctly recognize. Expand
Hunting for Wolves in Speaker Recognition
This work aims to predict which speaker pairs will be difficult for automatic speaker recognition systems to distinguish, by using features that characterize speakers, and thus provide a measure of speaker similarity, using data from NIST's 2008 Speaker Recognition Evaluation. Expand
Speaker model selection based on the Bayesian information criterion applied to unsupervised speaker indexing
This work proposes a flexible framework in which an optimal speaker model (GMM or VQ) is automatically selected based on the Bayesian Information Criterion according to the amount of training data available, and demonstrates that speaker indexing with this framework is sufficiently accurate for adaptation of the acoustic model. Expand
Speaker Recognition for DSR
Due to the coexistence of different compression algorithms in the fixed and mobile telephone networks, it is impossible to predict which combination of coders and channels the speech has undergoneExpand
Class-Discriminative Weighted Distortion Measure for VQ-based Speaker Identification
A weighted distortion measure is introduced that takes into account the correlations between the known models in the speaker database and larger weights are assigned to vectors that have high discriminating power between the speakers and vice versa. Expand
Ten Experiments on the Modeling of Polyphonic Timbre. (Dix Expériences sur la Modélisation du Timbre Polyphonique)
This thesis constructs an explicit measure of the timbre similarity between polyphonic music textures, and variants thereof inspired by previous work in Music Information Retrieval and shows that the precision of such measures is bounded, and that the remaining error rate is not incidental. Expand


A vector quantization approach to speaker recognition
A vector quantization (VQ) codebook was used as an efficient means of characterizing the short-time spectral features of a speaker and was used to recognize the identity of an unknown speaker from his/her unlabelled spoken utterances based on a minimum distance (distortion) classification rule. Expand
SHEEP, GOATS, LAMBS and WOLVES: a statistical analysis of speaker performance in the NIST 1998 speaker recognition evaluation
This paper proposes statistical tests for the existence of sheep, goats, lambs and wolves and applies these tests to hunt for such animals using results from the 1998 NIST speaker recognition evaluation. Expand
On the use of instantaneous and transitional spectral information in speaker recognition
  • F. Soong, A. Rosenberg
  • Computer Science
  • ICASSP '86. IEEE International Conference on Acoustics, Speech, and Signal Processing
  • 1986
The experimental results show that the instantaneous and transitional representations are relatively uncorrelated thus providing complementary information for speaker recognition, and simple transmission channel variations are shown to affect the instantaneous spectral representations and the corresponding recognition performance significantly, while the transitional representations and performance are relatively resistant. Expand
Evaluation of a vector quantization talker recognition system in text independent and text dependent modes
Although the approach is intrinsically text-independent, the system can be easily extended to text-dependent operation for improved performance and security by encoding specified training word utterances to form word prototypes. Expand
A new codebook training algorithm for VQ-based speaker recognition
  • J. He, Li Liu, G. Palm
  • Computer Science
  • 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing
  • 1997
A heuristic training procedure is proposed to retrain the codebooks so that they give a lower classification error rate for randomly selected vector-groups in VQ-based speaker recognition. Expand
Text-Independent Speaker Identification
The vector quantization approach will be used, due to ease of implementation and high accuracy, in this project. Expand
Vector quantization and signal compression
  • A. Gersho, R. Gray
  • Mathematics, Computer Science
  • The Kluwer international series in engineering and computer science
  • 1991
The author explains the design and implementation of the Levinson-Durbin Algorithm, which automates the very labor-intensive and therefore time-heavy and expensive process of designing and implementing a Quantizer. Expand
Gish and M . Schmit , “ Text - Independent Speaker Identification ”
  • Vector Quantization and Signal Compression ”
  • 1986