CONF astrid-00-21b/IDIAP Comparison of HMM experts with MLP experts in the Full Combination Multi-Band Approach to Robust ASR Hagen, Astrid Morris, Andrew full combination HMM/ANN-Hybrid multi-band multi-stream EXTERNAL https://publications.idiap.ch/attachments/reports/2000/rr00-21.pdf PUBLIC https://publications.idiap.ch/index.php/publications/showcite/astrid-00-21a Related documents ICSLP 2000 IDIAP-RR 00-21 In this paper we apply the Full Combination (FC) multi-band approach, which has originally been introduced in the framework of posterior-based HMM/ANN (Hidden Markov Model/Artificial Neural Network) hybrid systems, to systems in which the ANN (or Multilayer Perceptron (MLP)) is itself replaced by a Multi Gaussian HMM (MGM). Both systems represent the most widely used statistical models for robust ASR (automatic speech recognition). It is shown how the FC formula for the likelihood--based MGMs can easily be derived from the posterior-based approach by simply applying Bayes' Rule. The experiments show that the Full Combination multi-band system with MGM experts performs better, in all noise conditions tested, than the simple sum and product rules which are normally used. As compared to the baseline full-band system, the FC system shows increased robustness mainly on band-limited noise. The goal of this article is not a performance comparison between Multilayer Perceptrons and Multi Gaussian Models but between the theory of the two approaches, posterior-based vs. likelihood-based FC approach, so results are only given for the MGMs. REPORT astrid-00-21a/IDIAP Comparison of HMM experts with MLP experts in the Full Combination Multi-Band Approach to Robust ASR Hagen, Astrid Morris, Andrew full combination HMM/ANN-Hybrid multi-band multi-stream EXTERNAL https://publications.idiap.ch/attachments/reports/2000/rr00-21.pdf PUBLIC Idiap-RR-21-2000 2000 IDIAP Martigny, Switzerland July 2000 Published: ICSLP 2000, Beijing, September 2000 In this paper we apply the Full Combination (FC) multi-band approach, which has originally been introduced in the framework of posterior-based HMM/ANN (Hidden Markov Model/Artificial Neural Network) hybrid systems, to systems in which the ANN (or Multilayer Perceptron (MLP)) is itself replaced by a Multi Gaussian HMM (MGM). Both systems represent the most widely used statistical models for robust ASR (automatic speech recognition). It is shown how the FC formula for the likelihood-based MGMs can easily be derived from the posterior-based approach by simply applying Bayes' Rule. The experiments show that the Full Combination multi-band system with MGM experts performs better, in all noise conditions tested, than the simple sum and product rules which are normally used. As compared to the baseline full-band system, the FC system shows increased robustness mainly on band-limited noise. The goal of this article is not a performance comparison between Multilayer Perceptrons and Multi Gaussian Models but between the theory of the two approaches, posterior-based vs. likelihood-based FC approach, so results are only given for the MGMs.