PROSODY PREDICTION FOR TAMIL TEXT-TO-SPEECH SYNTHESIZER USING SENTIMENT ANALYSIS
DOI:
https://doi.org/10.22159/ajpcr.2017.v10s1.19535Keywords:
Natural language processing, Prosody, Sentiment analysis, Tamil, Text-to-speechAbstract
A speech synthesizer which sounds similar to a human voice is preferred over a robotic voice, and hence to increase the naturalness of a speech synthesizer an efficacious prosody model is imperative. Hence, this paper is focused on developing a prosody prediction model using sentiment analysis for a Tamil speech synthesizer. Two variations of prosody prediction models using SentiWordNet are experimented: one without a stemmer and the other with a stemmer. The prosody prediction model with a stemmer performs much more efficiently than the one without a stemmer as it tackles the highly agglutinative and inflectional words in Tamil language in a better way and is exemplified clearly, in this paper. The performance of the prosody prediction model with a stemmer has a higher classification accuracy of 77% on the test set in comparison to the 57% accuracy by the prosody model without a stemmer.
Â
Downloads
References
Bellur A, Narayan KB, Krishnan KR, Murthy HA. Prosody Modeling for Syllable-Based Concatenative Speech Synthesis of Hindi and Tamil. In: Communications (NCC), 2011 National Conference on 2011 January 28, IEEE. p. 1-5.
Reddy VR, Rao KS. Prosody modeling for syllable based text-tospeech synthesis using feedforward neural networks. Neurocomputing 2016;171:1323-34.
Rao KS, Yegnanarayana B. Modeling durations of syllables using neural networks. Comput Speech Lang 2007;21(2):282-95.
Rao KS, Koolagudi SG. Selection of suitable features for modeling the durations of syllables. J Softw Eng Appl 2010;3(12):1107.
Rao KS, Yegnanarayana B. Intonation modeling for Indian languages.
Comput Speech Lang 2009;23(2):240-56.
Reddy VR, Rao KS. Two-stage intonation modeling using feedforward neural networks for syllable based text-to-speech synthesis. Comput Speech Lang 2013;27(5):1105-26.
Rajendran V, Kumar GB. Text processing for developing unrestricted Tamil text to speech synthesis system. Indian J Sci Technol 2015;8(29):1-10.
Das A, Bandyopadhyay S. Sentiwordnet for Indian Languages. China: Asian Federation for Natural Language Processing; 2010. p. 56-63.
Antony JB, Mahalakshmi GS. Challenges in morphological analysis of tamil biomedical texts. Indian J Sci Technol 2015;8(23):1.
Tamil Stemmer. Available from: http://www.github.com/rdamodharan/ tamil-stemmer.
Kumar MA, Dhanalakshmi V, Soman KP, Rajendran S. A sequence labeling approach to morphological analyzer for Tamil language. Int J Comput Sci Eng 2010;2(6):1944-5.
Kumar MA, Dhanalakshmi V, Rekha RU, Soman KP, Rajendran S. Anovel data driven algorithm for Tamil morphological generator. Int J Comput Appl 2010;6(12):52-6.
Published
How to Cite
Issue
Section
The publication is licensed under CC By and is open access. Copyright is with author and allowed to retain publishing rights without restrictions.
