An Optimal Unit-Selection Algorithm for Ultra Low Bit-Rate Speech Coding

Ramasubramanian, V.; Harish, D.

doi:10.1109/icassp.2007.366969

Cited by 9 publications

(16 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A major part of this book will indeed focus on such unit-selection frameworks (Chaps. 3-6), starting with the pioneering work of Lee and Cox [LC01,LC02], followed by our own contributions [RH06, RH07, HR08, RH08, RH09,R12] that are in the form of i. unified, optimal and generalized unit-selection algorithms [RH06], [RH07] ii. comparative study with vector quantization, matrix quantization and variable length segment quantization [HR08] iii.…”

Section: Speech-to-speech Synthesis By Unit-selectionmentioning

confidence: 99%

“…Chapter 4 is exclusively devoted to our initial work in the unit-selection based coding framework [RH06,RH07] wherein we present optimal and unified unit-selection algorithms based on the one-pass dynamic programming framework and study its relative rate-distortion advantages over the sub-optimal segmental version of Lee and Cox [LC02]. Further in this chapter, we also provide a comprehensive comparative study of how the unit-selection framework fares in terms of ratedistortion performance over the clustered codebook based vector quantization, matrix quantization and variable-length segment quantization frameworks [HR08].…”

Section: Organization Of the Bookmentioning

confidence: 99%

“…Chapter 6 provides an interesting and important contribution in the form of a joint spectral-residual quantization paradigm [RH09,R12], within the unitselection frameworks of Lee and Cox [LC02] and our one-pass DP framework [RH06], [RH07], which obviates the need to transmit any information about the residual in a LPC vocoding framework. This result is likely to redefine the way LPC vocoding is done, wherein a speech-to-speech synthesis (STS) framework becomes completely self-contained in finding units that can be used for speech synthesis at the decoder without needing any information about the residual.…”

Section: Organization Of the Bookmentioning

confidence: 99%

“…Here [RH06,RH07], we proposed a unified and generalized framework for segment quantization of speech at ultra low bit-rates of 150 bps based on unit-selection principle using a modified one-pass dynamic programming algorithm and showed how it is exactly optimal for both fixed and variable-length segments and how it solves the sub-optimality of the Lee and Cox [LC02] algorithm by performing unit-selection based quantization 'directly' using the units of a continuous codebook without pre-quantizing the input speech. The unified unit-selection algorithm proposed by us handles both fixed-and variablelength units in a unified manner, thereby providing a generalization over both the two unit selection methods of Lee and Cox [LC01,LC02] which deal with 'singleframe' and 'segmental' units in a disparate manner.…”

Section: Unit-selection Paradigmsmentioning

confidence: 99%

“…In these algorithms presented by us [RH06,RH07], we showed three important results with respect to the early algorithms of Lee and Cox [LC01,LC02], based on rate-distortion curves using a very large continuous speech multi-speaker database: Firstly, our algorithm has a significantly superior performance than the segmental algorithm of Lee and Cox [LC02], by achieving considerably lower spectral distortions (up to 1.5 dB less) as well as much lower bit-rates over a range of database sizes; secondly, fixed length units perform significantly better than singleframe units, an important aspect overlooked by Lee and Cox's algorithm; and thirdly, use of fixed length units of 6-8 frames length offer spectral distortions similar to that of variable-length phonetic units, thereby circumventing expensive segmentation and labeling (manual or even automatic) of a continuous database (to define the variable-length units) for unit selection based low bit-rate coding.…”

Section: Unit-selection Paradigmsmentioning

confidence: 99%

See 4 more Smart Citations

Ultra Low Bit-Rate Speech Coding

Ramasubramanian¹,

Doddala²

2015

SpringerBriefs in Electrical and Computer Engineering

Self Cite

View full text Add to dashboard Cite

The authors of this series have been hand-selected. They comprise some of the most outstanding scientists -drawn from academia and private industry -whose research is marked by its novelty, applicability, and practicality in providing broad based speech solutions. The SpringerBriefs in Speech Technology series provides the latest findings in speech technology gleaned from comprehensive literature reviews and empirical investigations that are performed in both laboratory and real life settings. Some of the topics covered in this series include the presentation of real life commercial deployment of spoken dialog systems, contemporary methods of speech parameterization, developments in information security for automated speech, forensic speaker recognition, use of sophisticated speech analytics in call centers, and an exploration of new methods of soft computing for improving human-computer interaction. Those in academia, the private sector, the self-service industry, law enforcement, and government intelligence, are among the principal audience for this series, which is designed to serve as an important and essential reference guide for speech developers, system designers, speech engineers, linguists, and others. In particular, a major audience of readers will consist of researchers and technical experts in the automated call center industry where speech processing is a key component to the functioning of customer care contact centers.Amy Neustein, Ph.D., serves as Editor-in-Chief of the International Journal of Speech Technology (Springer). She edited the recently published book "Advances in Speech Recognition: Mobile Environments, Call Centers and Clinics" (Springer 2010), and serves as quest columnist on speech processing for Womensenews. Dr. Neustein is Founder and CEO of Linguistic Technology Systems, a NJ-based think tank for intelligent design of advanced natural language based emotion-detection software to improve human response in monitoring recorded conversations of terror suspects and helpline calls. Dr. Neustein's work appears in the peer review literature and in industry and mass media publications. Her academic books, which cover a range of political, social and legal topics, have been cited in the Chronicles of Higher Education, and have won her a Pro Humanitate Literary Award. She serves on the visiting faculty of the National Judicial College and as a plenary speaker at conferences in artificial intelligence and computing. Dr. Neustein is a member of MIR (machine intelligence research) Labs, which does advanced work in computer technology to assist underdeveloped countries in improving their ability to cope with famine, disease/illness, and political and social affliction. She is a founding member of the New York City Speech Processing Consortium, a newly formed group of NY-based companies, publishing houses, and researchers dedicated to advancing speech technology research and development.

show abstract