A new keyword spotting approach based on reward function.
Yassine Ben AyedDominique FohrJean Paul HatonGérard CholletPublished in: ISSPA (1) (2003)
Keyphrases
- keyword spotting
- reward function
- markov decision processes
- speech recognition
- reinforcement learning
- state space
- hidden markov models
- inverse reinforcement learning
- optimal policy
- multiple agents
- speech processing
- printed documents
- transition probabilities
- generative model
- state variables
- initially unknown
- multi agent
- handwritten documents
- search algorithm
- dynamic programming
- multimedia systems
- language model
- text mining
- information extraction