Abstract
This chapter first introduces the transmission channels employed currently for speech communication and their main impairments and then presents the literature review, divided into three parts: channel quality evaluation, human speaker recognition, and automatic speaker recognition. Different procedures for evaluation and main outcomes relevant to this work are indicated. The review of channel quality evaluation reports the current status of investigations addressing subjective perceptions and automatic evaluations of signal quality when the speech is transmitted through different kinds of communication channels. The rest of this review shows state-of-the-art methods to assess the human and the automatic speaker recognition performances, and the channel impairment effects that have been reported in previous investigations. On the human side, pertinent listening tests to assess the human capability to detect speaker identities reveal how the performance is influenced by different voice distortions. On the automatic side, a review of the most recent and efficient methods for automatic speaker recognition and their main findings under channel degradations are presented. Based on the fact that channels of extended bandwidths generally offer better quality and on the assessed importance of different speech frequency ranges for speaker recognition, this book concentrates on evaluating the advantages of enhanced channels for the human and for the automatic speaker recognition performance, clarifying how transmissions affect the speaker-specific voice properties and their relation to signal quality measurements.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The results of the NIST SRE 2012 challenge are reported in http://www.nist.gov/itl/iad/mig/sre12results.cfm, last accessed 28th September 2014.
- 2.
Already computed i-vectors were provided in the NIST 2014 Machine Learning Challenge with the aim of involving the machine learning community in the speaker recognition task [94].
- 3.
The Domain Adaptation Challenge (DAC) was organised in the summer 2013 by the Johns Hopkins University (JHU). The challenge description is given in http://www.clsp.jhu.edu/user_uploads/workshops/ws13/DAC_description_v2.pdf, last accessed 28th September 2014.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2016 Springer Science+Business Media Singapore
About this chapter
Cite this chapter
Fernández Gallardo, L. (2016). Literature Review. In: Human and Automatic Speaker Recognition over Telecommunication Channels. T-Labs Series in Telecommunication Services. Springer, Singapore. https://doi.org/10.1007/978-981-287-727-7_2
Download citation
DOI: https://doi.org/10.1007/978-981-287-727-7_2
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-287-726-0
Online ISBN: 978-981-287-727-7
eBook Packages: EngineeringEngineering (R0)