Consideration of Lombard effect for speechreading
| Autor: | Tsuhan Chen, Fu Jie Huang |
|---|---|
| Rok vydání: | 2002 |
| Předmět: |
Speechreading
Audio mining Computer science business.industry Speech recognition Speech coding Acoustic model Computer Science::Computation and Language (Computational Linguistics and Natural Language and Speech Processing) Pattern recognition Data_CODINGANDINFORMATIONTHEORY Viterbi algorithm Speech processing Lombard effect symbols.namesake Computer Science::Sound symbols Artificial intelligence Hidden Markov model business Computer Science::Information Theory |
| Zdroj: | MMSP |
| DOI: | 10.1109/mmsp.2001.962800 |
| Popis: | We propose a method for integrating audio and visual information to enhance speech recognition in adverse environments. We train the audio hidden Markov model and the visual hidden Markov model separately, and then use a Viterbi algorithm to decode both channels in parallel. The decoding process is asynchronous between the two channels to capture the asynchronous nature of audio and visual speech. We test the proposed method using speech corrupted by various types of noise and speech with the Lombard effect. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|