融合多任務學習類神經網路聲學模型訓練於會議語音辨識之研究(Leveraging Multi-task Learning with Neural Network Based Acoustic Modeling for Improved Meeting Speech Recognition) [In Chinese].
Ming-Han YangYao-Chi HsuHsiao-Tsung HungYing-Wen ChenBerlin ChenKuan-Yu ChenPublished in: ROCLING (2016)
Keyphrases
- speech recognition
- multi task learning
- speech recognition systems
- speech synthesis
- speaker independent
- multi task
- speech recognizers
- hidden markov models
- acoustic models
- speech signal
- gaussian processes
- neural network
- language model
- pattern recognition
- speech recognizer
- speaker diarization
- learning tasks
- automatic speech recognition
- learning problems
- high order
- transfer learning
- speaker identification
- learning algorithm
- acoustic features
- text mining
- information retrieval