Bayesian Subspace Hidden Markov Model for Acoustic Unit Discovery
Autor: | Jan Cernocký, Hari Krishna Vydana, Lukas Burget, Lucas Ondel |
---|---|
Přispěvatelé: | Faculty of Information Technology [Brno] (FIT / BUT), Brno University of Technology [Brno] (BUT) |
Rok vydání: | 2019 |
Předmět: |
FOS: Computer and information sciences
Sound (cs.SD) Computer Science - Machine Learning Computer science Speech recognition Bayesian probability Word error rate Machine Learning (stat.ML) TIMIT [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL] Computer Science - Sound Machine Learning (cs.LG) Set (abstract data type) 030507 speech-language pathology & audiology 03 medical and health sciences [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG] Audio and Speech Processing (eess.AS) Statistics - Machine Learning FOS: Electrical engineering electronic engineering information engineering Hidden Markov model ComputingMilieux_MISCELLANEOUS Subspace Gaussian Mixture Model Computer Science::Computation and Language (Computational Linguistics and Natural Language and Speech Processing) Computer Science::Sound Embedding 0305 other medical science Subspace topology Electrical Engineering and Systems Science - Audio and Speech Processing |
Zdroj: | INTERSPEECH INTERSPEECH, 2019, Graz, Austria |
Popis: | This work tackles the problem of learning a set of language specific acoustic units from unlabeled speech recordings given a set of labeled recordings from other languages. Our approach may be described by the following two steps procedure: first the model learns the notion of acoustic units from the labelled data and then the model uses its knowledge to find new acoustic units on the target language. We implement this process with the Bayesian Subspace Hidden Markov Model (SHMM), a model akin to the Subspace Gaussian Mixture Model (SGMM) where each low dimensional embedding represents an acoustic unit rather than just a HMM's state. The subspace is trained on 3 languages from the GlobalPhone corpus (German, Polish and Spanish) and the AUs are discovered on the TIMIT corpus. Results, measured in equivalent Phone Error Rate, show that this approach significantly outperforms previous HMM based acoustic units discovery systems and compares favorably with the Variational Auto Encoder-HMM. Accepted to Interspeech 2019 * corrected typos * Recalculated the segmentation using +-2 frames tolerance to comply with other publications |
Databáze: | OpenAIRE |
Externí odkaz: |