A Survey on Automatic Speaker Recognition Systems

Saquib, Zia; Salam, Nirmala; Nair, Rekha; Pandey, Nipun; Joshi, Akanksha

doi:10.1007/978-3-642-17641-8_18

Cited by 31 publications

(11 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Audio analysis has been traditionally focused on the recognition of speech [1,3], speaker identification [6,10,26,27] and scene categorization [4,22]. Recently, research in the area of intelligent surveillance systems shifted its attention to the automatic detection of abnormal or dangerous events through the analysis of audio streams acquired by microphones.…”

Section: Introductionmentioning

confidence: 99%

Reliable detection of audio events in highly noisy environments

Foggia

Petkov

Saggese

et al. 2015

Pattern Recognition Letters

142

115

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

Reliable detection of audio events in highly noisy environments

Foggia

Petkov

Saggese

et al. 2015

Pattern Recognition Letters

142

115

View full text Add to dashboard Cite

“…where x n is the original speech signal, w n is the Hamming window function, and k corresponds to the frequency. A power spectrogram can be converted into a log-power representation using (2).…”

Section: B Experimental Setup 1) Selected Features (Spectrograms)mentioning

confidence: 99%

“…Identifying a person by their voice is an important human trait that is typically taken for granted in natural human-tohuman interactions/communications. Human speech, which is a performance biometric, is different from other kinds of biometrics (such as hand geometry or fingerprints) [1] in that voice biometrics are the only commercial biometric product that process acoustic information [2]. A speaker's identity is represented in the way they speak, but not necessarily in the words being spoken [1].…”

Section: Introductionmentioning

confidence: 99%

Speaker Identification in Different Emotional States in Arabic and English

et al. 2020

View full text Add to dashboard Cite

Speaker recognition is an important application of digital speech processing. However, a major challenge degrading the robustness of speaker-recognition systems is variation in the emotional states of speakers, such as happiness, anger, sadness, or surprise. In this paper, we propose a speaker recognition system corresponding to three states, namely emotional, neutral, and with no consideration for a speaker's state (i.e., the speaker can be in an emotional state or neutral state), for two languages: Arabic and English. Additionally, cross-language speaker recognition was applied in emotional, neutral, and (emotional + neutral) states. Convolutional neural network and long short-term memory models were used to design a convolutional recurrent neural network (CRNN) main system. We also investigated the use of linearly spaced spectrograms as speech-feature inputs. The proposed system utilizes the KSUEmotions, emotional prosody speech and transcripts, WEST POINT, and TIMIT corpora. The CRNN system exhibited accuracies as high as 97.4% and 97.18% for Arabic and English emotional speech inputs, respectively, and 99.89% and 99.4% for Arabic and English neutral speech inputs, respectively. For the cross-language program, the overall CRNN system accuracy was as high as 91.83%, 99.88%, and 95.36% for emotional, neutral, and (emotional + neutral) states, respectively.

show abstract

“…The specific voice features refer to various analysis such as amplitude spectrum, localization of spectral peaks related to the vocal tract shape or pitch striations related to the user's glottal source. Similar to keystroke analysis, the speaker recognition can be performed either in static (text-dependent) or dynamic mode (text-independent) mode [29]. In the text-dependent mode the use is asked by the biometric system to pronounce a particular phrase, while, in the case of the text-independent mode the user is free to speak any phrase.…”

Section: Speaker Recognitionmentioning

confidence: 99%

Biometrics Systems and Technologies: A survey

Buciu

Gacsádi

2016

INT J COMPUT COMMUN

View full text Add to dashboard Cite

In a nutshell, a biometric security system requires a user to provide some biometric features which are then verified against some stored biometric templates. Nowadays, the traditional password based authentication method tends to be replaced by advanced biometrics technologies. Biometric based authentication is becoming increasingly appealing and common for most of the human-computer interaction devices. To give only one recent example, Microsoft augmented its brand new Windows 10 OS version with the capability of supporting face recognition when the user login in. This chapter does not intend to cover a comprehensive and detailed list of biometric techniques. The chapter rather aims at briefly discussing biometric related items, including principles, definitions, biometric modalities and technologies along with their advantages, disadvantages or limitations, and biometric standards, targeting unfamiliar readers. It also mentions the attributes of a biometric system as well as attacks on biometrics. Important reference sources are pointed out so that the interested reader may gain deeper in-depth knowledge by consulting them.

show abstract

A Survey on Automatic Speaker Recognition Systems

Cited by 31 publications

References 17 publications

Reliable detection of audio events in highly noisy environments

Reliable detection of audio events in highly noisy environments

Speaker Identification in Different Emotional States in Arabic and English

Biometrics Systems and Technologies: A survey

Contact Info

Product

Resources

About