Speech Communication and Signal Processing

B YEGNANARAYANA

doi:10.1007/s12046-011-0037-1

Helmholtz-Rayleigh Interdisciplinary Silver Medal in Psychological and Physiological Acoustics, Speech Communication, and Signal Processing in Acoustics: Blake S. Wilson

The Journal of the Acoustical Society of America ◽

10.1121/1.4988284 ◽

2017 ◽

Vol 141 (5) ◽

pp. 3761-3764

Author(s):

Michael Dorman ◽

Fan Gang Zeng ◽

John Hansen

Keyword(s):

Signal Processing ◽

Speech Communication ◽

Physiological Acoustics ◽

Silver Medal

Download Full-text

Speech Technology Progress Based on New Machine Learning Paradigm

Computational Intelligence and Neuroscience ◽

10.1155/2019/4368036 ◽

2019 ◽

Vol 2019 ◽

pp. 1-19 ◽

Cited By ~ 5

Author(s):

Vlado Delić ◽

Zoran Perić ◽

Milan Sečujski ◽

Nikša Jakovljević ◽

Jelena Nikolić ◽

...

Keyword(s):

Machine Learning ◽

Signal Processing ◽

Speech Signal ◽

Speech Synthesis ◽

Machine Learning Algorithms ◽

Speech Communication ◽

Speech Signal Processing ◽

Speech Technology ◽

Spoken Dialogue Systems ◽

New Machine

Speech technologies have been developed for decades as a typical signal processing area, while the last decade has brought a huge progress based on new machine learning paradigms. Owing not only to their intrinsic complexity but also to their relation with cognitive sciences, speech technologies are now viewed as a prime example of interdisciplinary knowledge area. This review article on speech signal analysis and processing, corresponding machine learning algorithms, and applied computational intelligence aims to give an insight into several fields, covering speech production and auditory perception, cognitive aspects of speech communication and language understanding, both speech recognition and text-to-speech synthesis in more details, and consequently the main directions in development of spoken dialogue systems. Additionally, the article discusses the concepts and recent advances in speech signal compression, coding, and transmission, including cognitive speech coding. To conclude, the main intention of this article is to highlight recent achievements and challenges based on new machine learning paradigms that, over the last decade, had an immense impact in the field of speech signal processing.

Download Full-text

Common Sound Scenarios: A Context-Driven Categorization of Everyday Sound Environments for Application in Hearing-Device Research

Journal of the American Academy of Audiology ◽

10.3766/jaaa.15105 ◽

2016 ◽

Vol 27 (07) ◽

pp. 527-540 ◽

Cited By ~ 15

Author(s):

Florian Wolters ◽

Karolina Smeds ◽

Erik Schmidt ◽

Eva Kümmel Christensen ◽

Christian Norup

Keyword(s):

Signal Processing ◽

Laboratory Tests ◽

Western Europe ◽

Literature Search ◽

Real Life ◽

Field Trials ◽

Normal Hearing ◽

Speech Communication ◽

Hearing Device ◽

Acoustic Environments

Background: Evaluation of hearing-device signal-processing features is performed for research and development purposes, but also in clinical settings. Most people agree that the benefit experienced in a hearing-device user’s daily life is most important, but laboratory tests are popular since they can be performed uniformly for all participants in a study using sensitive outcome measures. In order to design laboratory tests that have the potential of indicating real-life benefit, there is a need for more information about the acoustic environments and listening situations encountered by hearing-device users as well as by normal-hearing people. Purpose: To investigate the acoustic environments and listening situations people encounter, and to provide a structured framework of common sound scenarios (CoSS) that can be used for instance when designing realistic laboratory tests. Research Design: A literature search was conducted. Extracted acoustic environments and listening situations were categorized using a context-based approach. A set of common sound scenarios was established based on the findings from the literature. Data Collection: A number of publications providing data on encountered acoustic environments and listening situations were identified. Focus was on studies including informants who reported or recorded information in field trials. Nine relevant references were found. In combination with data collected at our laboratory, 187 examples of acoustic environments or listening situations were found. Results: Based on the extracted data, a categorization approach based on context (intentions and tasks) was used when creating CoSS. Three intention categories, “speech communication,” “focused listening,” and “nonspecific” were divided into seven task categories. In each task category, two sound scenarios were described, creating in total 14 common sound scenarios. The literature search showed a general lack of studies investigating acoustic environments and listening situations, in particular studies where normal-hearing informants are included and studies performed outside North America and Western Europe. Conclusions: A structured framework was developed. Intentions and tasks constitute the main categories in the framework, and 14 common sound scenarios were selected and described. The framework can for instance be used when developing hearing-device signal-processing features, in the evaluation of such features in realistic laboratory tests, and for demonstration of feature effects to hearing-device wearers.

Download Full-text

Speech Watermarking Method Using McAdams Coefficient Based on Random Forest Learning

Entropy ◽

10.3390/e23101246 ◽

2021 ◽

Vol 23 (10) ◽

pp. 1246

Author(s):

Candy Olivia Mawalim ◽

Masashi Unoki

Keyword(s):

Signal Processing ◽

Random Forest ◽

Communication Systems ◽

Objective Evaluation ◽

Random Forest Classifier ◽

Speech Communication ◽

Blind Detection ◽

Speech Watermarking ◽

Promising Solution

Speech watermarking has become a promising solution for protecting the security of speech communication systems. We propose a speech watermarking method that uses the McAdams coefficient, which is commonly used for frequency harmonics adjustment. The embedding process was conducted, using bit-inverse shifting. We also developed a random forest classifier, using features related to frequency harmonics for blind detection. An objective evaluation was conducted to analyze the performance of our method in terms of the inaudibility and robustness requirements. The results indicate that our method satisfies the speech watermarking requirements with a 16 bps payload under normal conditions and numerous non-malicious signal processing operations, e.g., conversion to Ogg or MP4 format.

Download Full-text