In the thesis, we showed that in the proposed framework: 1 the subword units can be graphemes, the units of written language, which make the pronunciation dictionary development easy; 2 the acoustic model can be trained on domain-independent or language-independent resources; and 3 the lexical model can be trained on a relatively small amount of transcribed speech data from the target domain or language in which we are interested to build an ASR system. The proposed approach facilitates sharing of resources and models from resource-rich languages and requires fewer or even zero conventional resources from the target language.
The potential and the efficacy of the proposed approach is demonstrated through experiments and comparisons with other standard approaches on ASR for resource rich languages, non-native and accented speech, under-resourced languages, and minority languages.
Tony Robinson (speech recognition)
The studies revealed that the proposed framework is particularly suitable when the task is challenged by the lack of both pronunciation dictionary and sufficient transcribed speech data. Furthermore, the investigations also showed that standard ASR approaches in which the lexical model is deterministic are more suitable when a phone-based pronunciation dictionary is available than for a grapheme-based pronunciation dictionary, while the probabilistic lexical model based ASR approach proposed in the thesis is suitable for both.
Automatic speech recognition; Kullback-Leibler divergence based hiddenMarkov model; lexicon; grapheme subword units; phoneme subword units; probabilistic lexical modeling; grapheme-based automatic speech recognition; grapheme-to-phoneme conversion; under-resourced speech recognition. Artificial Intelligence for Society.
Congratulations to her. Raj and R. Kim, W. Lim, and R. Kim, R. Stern, and H.
Eurospeech September, , Lisbon, Portugal. Raj, M. Seltzer, and R. Seltzer, B. Divenyi, Ed. Obuchi, N. Hataoka, and R. Li and R. Raj, R. Singh, and R. Seltzer, and B. Eurospeech September, , Geneva, Switzerland. Nedel and R. Obuchi and R. Singh, B. Singh, R. CRC Press, IEEE Conf. Li, R. Huerta and R. Singh, M. Ellis, R.
Singh, and S. Seltzer and B. Eurospeech September, , Aalborg, Denmark.
Doh and R. Nedel, R. Ravishankar, R. Eurospeech, September, , Budapest, Hungary. Huerta, S. Moreno, B. Stern, B. Raj, and P. Moreno, Siegler, U. Jain, B. Huerta, E. Thayer, M. Ravishankar, and R. Raj, E.
Raj, V. Parikh, and R. Eurospeech, September, , Rhodes, Greece. Stern, A. Acero, F. Liu, and Y. Lee and F. Soong, Eds.
Moreno, and R. Gouvea, P. Raj, T. Sullivan, and R. Pallett, Ed. Jain, M. Siegler, S.follow url
Computational neuroscience of speech recognition - ORA - Oxford University Research Archive
Doh, E. Siegler, and R. Moreno, M. Jain, and R. Moreno, U. III, pp. Stern, R. Liu, R. Acero, and P. Stern, F. Liu, P.