Acoustic Model Optimization Based on Evolutionary Stochastic Gradient Descent with Anchors for Automatic Speech Recognition.
Xiaodong CuiMichael PichenyPublished in: INTERSPEECH (2019)
Keyphrases
- stochastic gradient descent
- automatic speech recognition
- stochastic gradient
- speech recognition
- loss function
- least squares
- matrix factorization
- step size
- speech signal
- conversational speech
- hidden markov models
- support vector machine
- random forests
- regularization parameter
- weight vector
- genetic algorithm
- importance sampling
- online algorithms
- convergence rate
- cost function
- multi objective
- multiple kernel learning
- multiscale
- lower bound
- image processing