Badenhorst, JDavel, MH2016-08-222016-08-222015-11Badenhorst, J and Davel, MH. 2015. Synthetic triphones from trajectory-based feature distributions. In: Pattern Recognition Association of South Africa and Robotics and Mechatronics International Conference (PRASA-RobTech), Port Elizabeth, South Africa, 25-26 November 2015http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7359509&tag=1http://hdl.handle.net/10204/8737Pattern Recognition Association of South Africa and Robotics and Mechatronics International Conference (PRASA-RobTech), Port Elizabeth, South Africa, 25-26 November 2015We experiment with a new method to create synthetic models of rare and unseen triphones in order to supplement limited automatic speech recognition (ASR) training data. A trajectory model is used to characterise seen transitions at the spectral level, and these models are then used to create features for unseen or rare triphones. We find that a fairly restricted model (piece-wise linear with three line segments per channel of a diphone transition) is able to represent training data quite accurately. We report on initial results when creating additional triphones for a single-speaker data set, finding small but significant gains, especially when adding additional samples of rare (rather than unseen) triphones.enSynthetic triphonesTrajectory modellingTrajectory-based featuresFeature distributionsFeature constructionConference PresentationBadenhorst, J., & Davel, M. (2015). Synthetic triphones from trajectory-based feature distributions. IEEE. http://hdl.handle.net/10204/8737Badenhorst, J, and MH Davel. "Synthetic triphones from trajectory-based feature distributions." (2015): http://hdl.handle.net/10204/8737Badenhorst J, Davel M, Synthetic triphones from trajectory-based feature distributions; IEEE; 2015. http://hdl.handle.net/10204/8737 .TY - Conference Presentation AU - Badenhorst, J AU - Davel, MH AB - We experiment with a new method to create synthetic models of rare and unseen triphones in order to supplement limited automatic speech recognition (ASR) training data. A trajectory model is used to characterise seen transitions at the spectral level, and these models are then used to create features for unseen or rare triphones. We find that a fairly restricted model (piece-wise linear with three line segments per channel of a diphone transition) is able to represent training data quite accurately. We report on initial results when creating additional triphones for a single-speaker data set, finding small but significant gains, especially when adding additional samples of rare (rather than unseen) triphones. DA - 2015-11 DB - ResearchSpace DP - CSIR KW - Synthetic triphones KW - Trajectory modelling KW - Trajectory-based features KW - Feature distributions KW - Feature construction LK - https://researchspace.csir.co.za PY - 2015 T1 - Synthetic triphones from trajectory-based feature distributions TI - Synthetic triphones from trajectory-based feature distributions UR - http://hdl.handle.net/10204/8737 ER -