Pronunciation Rules in Portuguese Regional Speech (PORT REG) for Coarticulation Process
نویسندگان
چکیده
This paper describes one aspect of an ongoing work to incorporate pronunciation variability in the Portuguese (PORT) speech system. This work focuses on the linguistic rules to improve the grapheme-(multi)phone transcription algorithm that will be implemented. Portuguese ‘Beira Interior’ regional speech (PORT-BI REG) is considered to be in the realm of coarticulation (post-lexical) phenomena. A set of linguistic rules for most of the common vowel transformation in an utterance (vocalic segments at both the left and right edges of the word) is presented. The analysis focuses on the distinctive features that originate vowel sound challenges in connected speech. The results are interesting from the point of view of setting up models to reconstruct a grapheme-phone transcription algorithm for Portuguese multi-pronunciation speech systems. We propose that the linguistic documentation of Portuguese minority speech can be an optimal start for Portuguese speech system development process, too.
منابع مشابه
Generation of Word Pronunciation Networks from Automatically Learned Inter-word Coarticulation Rules
| In this paper a method for learning inter-word coarticulation rules from a training set is proposed. The algorithm is based on a comparison of the standard transcription (i.e. `how should it be pronounced') of each utterance with its auditory transcription (i.e. `what was actually pro-nounced'). It is shown that the latter transcriptions can be obtained without human intervention: the speech ...
متن کاملA Rule Based Pronunciation Generator and Regional Accent Databank for Portuguese
One of the major obstacles in deploying spoken language technologies (SLTs) in the developing world is a lack of key linguistic resources – e.g. electronic dictionaries, phonetically aligned corpora, pronunciation lexicons, etc. – that describe the non-dominant varieties spoken in such countries and regions. In this paper, we describe the work of the LUPo (Portuguese Unisyn Lexicon) project to ...
متن کاملA study of implicit and explicit modeling of coarticulation and pronunciation variation
In this paper, we focus on the modeling of coarticulation and pronunciation variation in Automatic Speech Recognition systems (ASR). Most ASR systems explicitly describe these production phenomena through context-dependent phoneme models and multiple pronunciation lexicons. Here, we explore the potential benefit of using feature spaces covering longer time segments in terms of implicit modeling...
متن کاملThis is a placeholder. Final title will be filled later
This paper describes some experiments with pronunciation modeling for spontaneous speech in European Portuguese. The transducer framework provides an elegant way to combine a pronunciation lexicon of canonical forms with alternative pronunciation rules. The main phonological aspects that the rules are intended to cover are: vowel devoicing, deletion and coalescence, voicing assimilation, and si...
متن کاملAllophone-based acoustic modeling for Persian phoneme recognition
Phoneme recognition is one of the fundamental phases of automatic speech recognition. Coarticulation which refers to the integration of sounds, is one of the important obstacles in phoneme recognition. In other words, each phone is influenced and changed by the characteristics of its neighbor phones, and coarticulation is responsible for most of these changes. The idea of modeling the effects o...
متن کامل