IEEE Signal Processing Magazine· 2012Q1
Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups
- 10,342citations
- Q1SCImago
- 2012year
Short summary
Deep neural networks (DNNs) with multiple hidden layers significantly outperform traditional Gaussian mixture models (GMMs) in acoustic modeling for speech recognition, achieving better benchmark results.
AI-generated from the title and abstract; the full text is not read.
Key points
- DNNs with many hidden layers outperform GMMs in speech recognition acoustic modeling.
- DNNs produce posterior probabilities over HMM states, an alternative to GMM-based state fitting.
- New training methods are crucial for DNN performance.
- This article consolidates the views of four leading research groups in the field.
AI-generated from the title and abstract; the full text is not read.
Abstract
Most current speech recognition systems use hidden Markov models (HMMs) to deal with the temporal variability of speech and Gaussian mixture models (GMMs) to determine how well each state of each HMM fits a frame or a short window of frames of coefficients that represents the acoustic input. An alternative way to evaluate the fit is to use a feed-forward neural network that takes several frames of coefficients as input and produces posterior probabilities over HMM states as output. Deep neural networks (DNNs) that have many hidden layers and are trained using new methods have been shown to outperform GMMs on a variety of speech recognition benchmarks, sometimes by a large margin. This article provides an overview of this progress and represents the shared views of four research groups that have had recent successes in using DNNs for acoustic modeling in speech recognition.
The authors' abstract, as published at the source. IEEE Signal Processing Magazine, 2012 · DOI ↗
Continue with a free account
Ask the paper: 3 free questions a day about this paper; save it, get its citation, new summaries every day for your field. Takeaways are Premium.
Continue free on the webSign in with Google or Apple; no card needed. You come back to this paper.
On your phone:
Field: Artificial Intelligence
Artificial IntelligenceComputer Science