Search

This paper presents a chat application that make use of speech as interaction mode between users in their mother tongue. This system is based on Machine Translation (MT) which automatically translates a sentence from one language to other. Now-a-days, Chat has become increasingly important as an effective means to communicate in social networks. But in existing personal chat systems, chat services are text-based only. To circumvent this issue, innovative, practical XMPP-based chat system architecture is proposed in this paper that creates a Speech based chat services by integrating state-of-the-art technologies like Automatic Speech Recognition (ASR), Text to Speech (TTS) and Machine Translation (MT). The proposed system eliminates the need to have a common language for chat communication. In this chat system the user speak in their own language after selecting the language of their comfort. The audio will be recognized in real time. The text is translated to the language opted by end user and the system synthesizes it. Here, we discuss an interface prototype that can overcome the language barrier to some extent.

Speech corpora are vital resource in development and evaluation of automatic speech recognition systems, as well as for acoustic phonetic studies. Collecting a huge corpus is not an easy task. The lack of such resources is one of the reasons for the absence of good quality speech recognition systems in Indian languages. Here we have automated such process by developing web based tool for collecting broad band speech data and an IVR system with speech recognition for collecting narrow band speech data. The main features includes the full support for the typical recording, annotation and project administration workflow, easy editing of the speech content, with an advantage of a fully localizable user interface. This paper describes in detail the development of web based speech collection tool and an IVR system which will enable end-to-end building of speech corpus with minimum manual effort.

Newspapers are the source of news and information and more over it's a window to the outside world Unfortunately some of us need others help to gain the knowledge of News Reading Voice Activated Multilingual Newspaper Reading System is a system which enables such people to get benefit of newspaper reading without depending others. It's an application which collects, classifies and reads online Newspapers in different languages. The system supports languages like Malayalam, Tamil, Hindi and English. The technologies like speech recognition, speech synthesis and the web are integrated together in the system. The major application of the system will be in the form of an assistive technology for visually or physically challenged, senior citizens and even the illiterate. The application also helps the busy personalities to read the Newspaper, as it provides hands free access to the news sites. The major enhancements are integration of Multilingual Automatic Speech Recognition (ASR), Multilingual Text-to-Speech system with other enhanced features like keyword search facility, Intelligent/ Auto customization in accordance with user and paper independent classified headings. The integration of ASR enables user to operate the system in complete hands free mode.

This paper describes the development of state-of-the-art large vocabulary continuous speech recognition (LVCSR) system for the Malayalam language with an application for visually challenged. For an LVCSR, building a high accurate acoustic models and large-scale language models are the challenging task. Speech corpus for training the system is collected from 80 native speakers in room environment ensuring the speaker variance. Mel-frequency Cepstral Coefficients (MFCC) method is used as a front-end to extract acoustic features from the input signal. Acoustic model is built on 30 hours of speech data based on Hidden Markov Model (HMM). A hybrid model, integrating rule based and statistical method is used to handle pronunciation variations in the dictionary. The best configuration of the system achieved word accuracy of 75% in average. Accuracy of the system is further increased up to 80% in average, by implementing speaker adaptation technique. The developed system is integrated to OpenOffice Writer together with TTS for making it user friendly editor for visually challenged people.

This paper describes the implementation of a dictation system for Malayalam office documents in OpenOffice Writer. Dictation system is built using state-of-the-art large vocabulary continuous speech recognition system for the Malayalam language. This system supports a vocabulary of 5000 most commonly used office domain words and is employed with a vocabulary updating facility to handle out-of-vocabulary words. The system is based on Hidden Markov Model (HMM), trained with huge (25 hours) amount of data. The training data is collected in room environment, ensuring the speaker variance and the phonetic richness. A hybrid model which integrates the rule based method with statistical method is used to handle the pronunciation variations for the creation of the pronunciation dictionary. The system is first of its kind which simplifies the tedious task of typing in Malayalam. Apart from dictating office documents with 75 ±5 % accuracy, the system is equipped with a facility of suggestion generation by which the user will be provided with alternate words for mis-recognized words. The system also supports some basic voice command operations for file operations like open, save, close etc. This system has an option to adapt to the user's voice which will improve the recognition accuracy by 2-5%. The system is successfully implemented in OpenOffice Writer and tested.

An accurate, properly labeled speech corpus is very important for speech research. However, manual segmentation and labeling is very laborious and error prone. This paper describes an automatic tool for segmenting and labeling of Malayalam speech data. The tool is based on Hidden Markov Model (HMM). HMM Tool Kit is used for training, segmentation and labeling the data. Special care was taken in the preparation of pronunciation dictionary so that it will cover most of the possible pronunciation variations. Syllabification rule is applied in the phone label for generating syllable label also.. Segmentation and labeling experiment was done on the speech corpus collected for building text-to-speech system. The performance of the tool is reasonably good as it shows only 19ms average deviation compared to manual labels.

This paper reports the results of a study conducted to determine the densities of Vibrio parahaemolyticus in market samples of finfish and shellfish from Cochin.

This brief focuses on the relationship between social economic status and the criminal justice system. In general, people pertaining to the lower class—versus the white collar and the elite class— are more likely to be: incarcerated, charged, convicted, sentenced to prison, and punished with longer prison terms.