Contribuições para o Atlas Prosódico Multimídia do Português do Norte do Brasil – AMPER-POR: variedade linguística do município de Abaetetuba (PA)

Abstract

The main goal of this Master Dissertation is to characterize the dialect prosodic variation of Brazilian Portuguese spoken in an Amazonian city, Abaetetuba. The methodology follows the AMPER project’s guidelines. The recordings have obtained at the fieldwork. The corpus consist of 102 sentences, SVC (Subject + Verb + Complement) and their expansions (Adjectival and Prepositional phrases) structured with the same phonetic and syntactic restrictions. As each sentence was repeating six times by each one of the 4 speakers, so the total corpus is compound by 612 phrases for each speaker. The pitch range was between 50 Hz and 250 Hz for males and 110 Hz to 370 Hz for females. Three controlled acoustic parameters are used: a) the fundamental frequency (F0), b) the duration (ms) and, c) the intensity (dB). The data analysis was taking through seven stages of treatment: 1) codification of repetitions, 2) isolation of sentences in individual audio files; 3) phonetic segmentation in software PRAAT; 4) application of PRAAT script; 5) selecting the three best utterances; 6) application of MATLAB interface; 7) It is used the EXCEL program to generate graphs used in the data analysis. The results show that “three major variations of acoustical parameters preferentially occur at the core element of the stressed syllable of the phrase and/ or last syllable of the utterance” (CRUZ & BRITO, 2011).