This paper describes the acoustic characteristics of Bangla consonants, obtained by analyzing the recordings of male and female voices. First, the duration of each phoneme was identified by averaging both the male and female voice data; then, formant were measured and formant comparison was made for controversial phonemes, which also served to resolve the controversies in the existing phoneme i...

This paper presents an approach for the generation and selection of pronunciation transcriptions for a exible word recognizer. The basic idea is to produce pronunciation variants and corresponding scores with a set of pronunciation variation rules, which are weighted with their frequencies of occurence measured on the training data. This approach addresses the problem of interfering transcripti...

developing reliable yet flexible software is a hard problem. although modeling methods enjoy a lot of advantages, the exclusive use of just one of them, in many cases, may not guarantee the development of reliable and flexible software. formal modeling methods ensure reliability because they use a rigorous approach to software development. however, lack of knowledge and high cost practically fo...

Phonetic alignment is the task of finding limits phones and higher units in an audio file. This has been reliably done many languages such as English, French German, but, so far, no available Brazilian Portuguese aligner had a performance comparable with ones used for these other languages. Thus, main goal this work was to implement useful tool forced Portuguese. The implementation two steps, g...

Multiple classifier fusion may generate more accurate classification than each of the constituent classifiers. Fusion is often based on fixed combination rules like the product and average. Only under strict probabilistic conditions can these rules be justified. We present here a simple rule for adapting the class combiner to the application. c decision templates (one per class) are estimated w...

Multiple classi"er fusion may generate more accurate classi"cation than each of the constituent classi"ers. Fusion is often based on "xed combination rules like the product and average. Only under strict probabilistic conditions can these rules be justi"ed. We present here a simple rule for adapting the class combiner to the application. c decision templates (one per class) are estimated with t...

Festival, a speech synthesis system developed by the University of Edinburgh, provides the framework for converting natural language into text-tospeech rules. Hawaiian language, with only 8 consonants and five vowels, can be implemented on the Festival system. However, Hawaiian’s special characters, the kahako and the okina, need special consideration which is not available in the Festival syst...

We describe an approach to grapheme to phoneme conver sion which is both language independent and data oriented Given a set of examples spelling words with their associated phonetic representation in a language a grapheme to phoneme conversion system is automatically pro duced for that language which takes as its input the spelling of words and produces as its output the phonetic transcription ...

We describe an approach to grapheme-to-phoneme conversion which is both language-independent and data-oriented. Given a set of examples (spelling words with their associated phonetic representation) in a language, a grapheme-to-phoneme conversion system is automatically produced for that language which takes as its input the spelling of words, and produces as its output the phonetic transcripti...