Prof. Hanna Bogucka, head of the Department of Wireless Communications at the Poznań University of Technology, discusses unnecessary inhibitions, the usefulness of microphones, and the links between people and technology.
As a consequence of recent implementations of EU Directives related to noise protection more and more students of various AGH-UST programs are introduced to the basics of acoustic measurements. Students at various levels of theoretical background in the field of acoustic measurements are offered practical training in measurements using digital sound analyzers. The situation would be optimal if each student could have a device at his/her own disposal. Unfortunately, such a situation is not possible at the moment because of various reasons.
With the above problem in mind, a dedicated software package has been developed, implemented in the LabVIEW environment, which allows detailed studies of problems related to the acoustic signal measurement using sound level meters, as well as tasks in spectral analysis (1/1 and 1/3 band filters) and narrow-band (FFT) analysis. With such organization during the introductory laboratory classes each student is offered a direct individual contact with a virtual device that is properly pre-programmed for realization of a well-constructed learning process. It definitely facilitates understanding of the essence of acoustic signal measurements and provides a good basis for further laboratory work carried out as a team-activity.
Research work on the design of robust multimodal speech recognition systems making use of acoustic and visual cues, extracted using the relatively noise robust alternate speech sensors is gaining interest in recent times among the speech processing research fraternity. The primary objective of this work is to study the exclusive influence of Lombard effect on the automatic recognition of the confusable syllabic consonant-vowel units of Hindi language, as a step towards building robust multimodal ASR systems in adverse environments in the context of Indian languages which are syllabic in nature. The dataset for this work comprises the confusable 145 consonant-vowel (CV) syllabic units of Hindi language recorded simultaneously using three modalities that capture the acoustic and visual speech cues, namely normal acoustic microphone (NM), throat microphone (TM) and a camera that captures the associated lip movements. The Lombard effect is induced by feeding crowd noise into the speaker’s headphone while recording. Convolutional Neural Network (CNN) models are built to categorise the CV units based on their place of articulation (POA), manner of articulation (MOA), and vowels (under clean and Lombard conditions). For validation purpose, corresponding Hidden Markov Models (HMM) are also built and tested. Unimodal Automatic Speech Recognition (ASR) systems built using each of the three speech cues from Lombard speech show a loss in recognition of MOA and vowels while POA gets a boost in all the systems due to Lombard effect. Combining the three complimentary speech cues to build bimodal and trimodal ASR systems shows that the recognition loss due to Lombard effect for MOA and vowels reduces compared to the unimodal systems, while the POA recognition is still better due to Lombard effect. A bimodal system is proposed using only alternate acoustic and visual cues which gives a better discrimination of the place and manner of articulation than even standard ASR system. Among the multimodal ASR systems studied, the proposed trimodal system based on Lombard speech gives the best recognition accuracy of 98%, 95%, and 76% for the vowels, MOA and POA, respectively, with an average improvement of 36% over the unimodal ASR systems and 9% improvement over the bimodal ASR systems.