Lexical access experiments with context-dependent articulatory feature-based models
| Autor: | Karen Livescu, Preethi Jyothi, Eric Fosler-Lussier |
|---|---|
| Rok vydání: | 2011 |
| Předmět: |
Context model
Computer science business.industry Speech recognition Decision tree Lexical access Context (language use) Pronunciation computer.software_genre Speech processing ComputingMethodologies_PATTERNRECOGNITION Variation (linguistics) Feature (machine learning) Artificial intelligence business computer Natural language processing Dynamic Bayesian network |
| Zdroj: | ICASSP |
| DOI: | 10.1109/icassp.2011.5947454 |
| Popis: | We address the problem of pronunciation variation in conversational speech with a context-dependent articulatory feature-based model. The model is an extension of previous work using dynamic Bayesian networks, which allow for easy factorization of a state into multiple variables representing the articulatory features. We build context-dependent decision trees for the articulatory feature distributions, which are incorporated into the dynamic Bayesian networks, and experiment with different sets of context variables. We evaluate our models on a lexical access task using a phonetically transcribed subset of the Switchboard corpus. We find that our models outperform a context-dependent phonetic baseline. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|