Novel Online Tools for Automatic Generation of Pronouncing Dictionaries in Mexican Spanish for Speech Processing
Carlos D. Hernández-Mena and Abel Herrera-Camacho
Signal Processing Department, National Autonomous University of Mexico (UNAM), Mexico City
Abstract—A pronouncing dictionary is a very important tool in a speech processing system. In speech recognition, it helps to the training stage to create the Markov models for every phoneme of every word in the lexicon. In speech synthesis, it helps the system to produce the correct pronunciations of the words introduced by the user despite the orthographic representations of them. All of this implies that creation of pronouncing dictionaries depends on the language one has chosen, because different languages have different sets of phonemes and also different phonetic alphabets to represent them. In this paper we present a methodology of creation of pronouncing dictionaries in Mexican Spanish utilizing a set of novel online tools developed by the CIEMPIESS-UNAM Project. These tools are for free use and they produce pronouncing dictionaries in a particular phonetic alphabet called Mexbet.
Index Terms—Mexican Spanish, Mexbet, pronouncing dictionary, phonetic alphabet
Cite: Carlos D. Hernández-Mena and Abel Herrera-Camacho, "Novel Online Tools for Automatic Generation of Pronouncing Dictionaries in Mexican Spanish for Speech Processing," International Journal of Signal Processing Systems, Vol. 4, No. 1, pp. 51-54, February 2016. doi: 10.12720/ijsps.4.1.51-54
Cite: Carlos D. Hernández-Mena and Abel Herrera-Camacho, "Novel Online Tools for Automatic Generation of Pronouncing Dictionaries in Mexican Spanish for Speech Processing," International Journal of Signal Processing Systems, Vol. 4, No. 1, pp. 51-54, February 2016. doi: 10.12720/ijsps.4.1.51-54