Speaker-aware Multi-Task Learning for automatic speech recognition.
Gueorgui PironkovStéphane DupontThierry DutoitPublished in: ICPR (2016)
Keyphrases
- automatic speech recognition
- multi task learning
- speech recognition
- multi task
- multiple tasks
- gaussian processes
- hidden markov models
- speech signal
- high order
- learning tasks
- acoustic features
- transfer learning
- broadcast news
- noisy environments
- learning problems
- theoretical analysis
- conversational speech
- spontaneous speech
- speech retrieval
- speech sounds
- feature selection
- language model
- multi class
- probabilistic model
- learning algorithm
- machine learning
- learning models
- non stationary
- semi supervised learning
- computer vision