A segment selection technique for speaker verification

Nosratighods, Mohaddeseh; Ambikairajah, Eliathamby; Epps, Julien; Carey, Michael J.

doi:10.1016/j.specom.2010.04.007

Cited by 18 publications

(6 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This work was extended in [6] which reported that the i-vector model can distill speaker information in a more effective way so it is more suitable for SUSR. In addition, a score-based segment selection technique was proposed in [7]. A relative EER reduction of 22% was reported by the authors on a recognition task where the test utterances were shorter than 15 seconds in length.…”

Section: Introductionmentioning

confidence: 99%

System combination for short utterance speaker recognition

Wang

Zhang

et al. 2016

2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA)

View full text Add to dashboard Cite

For text-independent short-utterance speaker recognition (SUSR), the performance often degrades dramatically. This paper presents a combination approach to the SUSR tasks with two phonetic-aware systems: one is the DNN-based ivector system and the other is our recently proposed subregionbased GMM-UBM system. The former employs phone posteriors to construct an i-vector model in which the shared statistics offers stronger robustness against limited test data, while the latter establishes a phone-dependent GMM-UBM system which represents speaker characteristics with more details. A score-level fusion is implemented to integrate the respective advantages from the two systems. Experimental results show that for the textindependent SUSR task, both the DNN-based i-vector system and the subregion-based GMM-UBM system outperform their respective baselines, and the score-level system combination delivers performance improvement.

show abstract

Section: Introductionmentioning

confidence: 99%

System combination for short utterance speaker recognition

Wang

Zhang

et al. 2016

2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA)

View full text Add to dashboard Cite

show abstract

“…Another work is the iVector approach and factor analysis subspace estimation introduced in [25,26] to reduce the number of redundant model parameters, resulting in more accurate speaker models. Some approaches attempt to increase performance by selecting segments with better discriminability based on speaker features [27] GMM and CNN hybrid method [28]. In their work, front-end feature extraction methods are based on Fourier transform Mel-triangle filtering and linear prediction cepstral coefficients for model training and testing as well as model inference.…”

Section: Related Workmentioning

confidence: 99%

Improving short-duration speaker recognition by joint Bark-wavelet acoustic feature coupling and triplet dual-attention mechanism network

Xiong

2023

Preprint

View full text Add to dashboard Cite

Speaker recognition methods are negatively affected by the input short duration audio signal. In this paper, we tackle the problem of speaker recognition from shorter speech data by coupling proposed two acoustic features: Bark-scaled Gauss Filter Cepstral Coefficients (BGCC) and Perceptual Wavelet Packet Entropy (PWPE). Our assumption is based on the observation that BGCC and PWPE capture sufficient information on various aspects of speech that can be used to discriminate speaker, viz., speech perception and high time-frequency information representation, etc., for enhancing characteristic diversity. A triplet dual attention mechanism (Triplet-DAM) is used to couple these two features in a creative manner. The coupling method means that the feature of limited short utterance can be reused, through the dual attention mechanism, more discriminative features are enhanced for limited feature, thus improving speaker recognition performance in short duration audio signals. Extensive analysis on a variety of datasets, which speech samples of different types, diverse lengths, etc., demonstrate the superiority of the proposed feature engineering and method over existing acoustic feature extraction and speaker recognition algorithms, including those based on MFCCs, LPCCs features, and GMM-UBM, iVector-PLDA, ResCNN-triplet, respectively. The experimental results demonstrate the proposed method achieves notable improvement with the existing approach for short duration speaker recognition.

show abstract

“…Besides, Nosratighods [104] presented a score-based segment selection technique for discarding portions of speech that results in poor discrimination ability in a speaker verification task. Theory is developed to detect the most significant and reliable speech segments based on the probability that test segment comes from a fixed set of cohort models.…”

Section: A To Select More Discriminative Datamentioning

confidence: 99%

An overview of robustness related issues in speaker recognition

Zheng¹,

Jin

Li³

et al. 2014

Signal and Information Processing Association Annual Summit and Conference (APSIPA), 2014 Asia-Pacific

View full text Add to dashboard Cite

Speaker recognition technologies have been improved rapidly in recent years. However, critical robustness issues need to be addressed when they are applied in practical situations. This paper provides an overview of technologies dealing with robustness related issues in automatic speaker recognition. We first categorize the robustness issues into three categories, including environment-related, speaker-related and application-oriented issues. For each category, we then describe the current hot topics, existing technologies, and potential research focuses in the future.I.

show abstract

A segment selection technique for speaker verification

Cited by 18 publications

References 10 publications

System combination for short utterance speaker recognition

System combination for short utterance speaker recognition

Improving short-duration speaker recognition by joint Bark-wavelet acoustic feature coupling and triplet dual-attention mechanism network

An overview of robustness related issues in speaker recognition

Contact Info

Product

Resources

About