Speaker recognition using HMM with experiments on the yoho database

Che, ChiWei; Lin, Qing

doi:10.21437/eurospeech.1995-104

Cited by 31 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is noted that least error rates are obtained when N is between 5 and 10. This result conforms with our previous speaker verication experiments on the YOHO database [9].…”

Section: Nist Spid Evaluationsupporting

confidence: 93%

Selective use of the speech spectrum and a VQGMM method for speaker identification

Lin¹,

Che²,

Yuk³

et al. 1996

4th International Conference on Spoken Language Processing (ICSLP 1996)

Self Cite

View full text Add to dashboard Cite

This paper describes two separate sets of speaker identication experiments. In the rst set of experiments, the speech spectrum is selectively used for speaker identication. The results show that the higher portion of the speech spectrum contains more reliable idiosyncratic information on speakers than does the lower portion of equal bandwidth. In the second set of experiments, a v ector-quantization based Gaussian mixture models (VQGMMs) is developed for text-independent speaker identication. The system has been evaluated in the recent speaker identication evaluation organized by NIST. In this paper, details of the system design are given and the evaluation results are presented.

show abstract

“…It is noted that least error rates are obtained when N is between 5 and 10. This result conforms with our previous speaker verication experiments on the YOHO database [9].…”

Section: Nist Spid Evaluationsupporting

confidence: 93%

Selective use of the speech spectrum and a VQGMM method for speaker identification

Lin¹,

Che²,

Yuk³

et al. 1996

4th International Conference on Spoken Language Processing (ICSLP 1996)

Self Cite

View full text Add to dashboard Cite

show abstract

“…The sentence was recorded live (direct) by using PRAAT 6.0 version software [9]. The sampling rate was selected at 44kHz.…”

Section: Methodsmentioning

confidence: 99%

Is tonal language a problem for speaker identification (SPID)?

P¹,

Vanlalhriati²

2022

World J. Adv. Eng. Technol. Sci.

View full text Add to dashboard Cite

The availability of numerous technologies has led to an increase in the usage of bioinformatics in recent years. Siri, Alexa, and other artificial intelligence systems assist us in our daily lives. Voice recognition systems are used to confirm an individual's identity based on particular elements retrieved from his or her voice. In this regard, the current study attempted to assess the proportion of speaker identification in tonal language speaking persons. The study included 20 participants in the age range from 20 to 40 years. All participants were given a few phrases to speak and were recorded. PRAAT software was used to analyze the obtained data. A vector was developed by using the first two formants, which was then utilized to calculate the percentage for speaker identification. From small sample size to bigger sample size, three groups were formed: A-5, B-10, and C-20 speakers. In a lower sample size, results showed a benchmark of 65% for vowel /i:/, which is better for SPID, 60% for /a:/, which is above chance level, and 45% for /u:/, which is below chance level. The authors stated that increasing the sample size has an influence on speaker identification.

show abstract