Van Niekerk, DRBarnard, ESchlunz, Georg I2010-01-082010-01-082009-11Van Niekerk, DR, Barnard, E and Schlunz, G. 2009. Perceptual evaluation of corpus-based speech synthesis techniques in under-resourced environments. 20th Annual Symposium of the Pattern Recognition Association of South Africa (PRASA). Stellenbosch, South Africa, 30 November - 01 December 2009, pp 71-75http://hdl.handle.net/10204/385220th Annual Symposium of the Pattern Recognition Association of South Africa (PRASA). Stellenbosch, South Africa, 30 November - 01 December 2009With the increasing prominence and maturity of corpus-based techniques for speech synthesis, the process of system development has in some ways been simplified considerably. However, the dependence on sufficient amounts of relevant speech data of high quality remains a central challenge in under-resourced environments. In this paper the authors investigate the quality implications when building baseline synthesis systems with reduced amounts of speech data. This is done through a perceptual evaluation of synthesis systems based on unit-selection and statistical parametric synthesis techniques. The authors show that - although it is possible to build an acceptable unit-selection synthesizer with as little as 27 minutes of carefully recorded speech data - synthesis quality obtainable from Hidden Markov Model-based synthesis is more consistent and requires significantly less speech data.enSpeech synthesis techniquesUnder-resourced environmentsPerceptual evaluationSpeech dataHidden markov modelsPRASA 2009Conference PresentationVan Niekerk, D., Barnard, E., & Schlunz, G. I. (2009). Perceptual evaluation of corpus-based speech synthesis techniques in under-resourced environments. PRASA 2009. http://hdl.handle.net/10204/3852Van Niekerk, DR, E Barnard, and Georg I Schlunz. "Perceptual evaluation of corpus-based speech synthesis techniques in under-resourced environments." (2009): http://hdl.handle.net/10204/3852Van Niekerk D, Barnard E, Schlunz GI, Perceptual evaluation of corpus-based speech synthesis techniques in under-resourced environments; PRASA 2009; 2009. http://hdl.handle.net/10204/3852 .TY - Conference Presentation AU - Van Niekerk, DR AU - Barnard, E AU - Schlunz, Georg I AB - With the increasing prominence and maturity of corpus-based techniques for speech synthesis, the process of system development has in some ways been simplified considerably. However, the dependence on sufficient amounts of relevant speech data of high quality remains a central challenge in under-resourced environments. In this paper the authors investigate the quality implications when building baseline synthesis systems with reduced amounts of speech data. This is done through a perceptual evaluation of synthesis systems based on unit-selection and statistical parametric synthesis techniques. The authors show that - although it is possible to build an acceptable unit-selection synthesizer with as little as 27 minutes of carefully recorded speech data - synthesis quality obtainable from Hidden Markov Model-based synthesis is more consistent and requires significantly less speech data. DA - 2009-11 DB - ResearchSpace DP - CSIR KW - Speech synthesis techniques KW - Under-resourced environments KW - Perceptual evaluation KW - Speech data KW - Hidden markov models KW - PRASA 2009 LK - https://researchspace.csir.co.za PY - 2009 T1 - Perceptual evaluation of corpus-based speech synthesis techniques in under-resourced environments TI - Perceptual evaluation of corpus-based speech synthesis techniques in under-resourced environments UR - http://hdl.handle.net/10204/3852 ER -