PORTUGUESE VARIETY IDENTIFICATION ON BROADCAST NEWS

This paper describes an accent identification system for Portuguese, that explores different type of properties: acoustic, phonotactic and prosodic. The system is designed to be used as a pre-processing module for the Portuguese Automatic Speech Recognition system developed at INESC-ID. In terms of variety identification, the over-all rate of correct identification is 69.0 % if all 7 varieties are consid-ered, and the best results are obtained for Brazilian Portuguese, also the variety that proved easiest to identify in perceptual experiments. When distinguishing between European, Brazilian and African Por-tuguese, the identification rate goes up to 94.7%. The fact that the prosodic system alone can achieve an identification rate of 77 % is also worth investigating