Efficient Keyword Spotting Using Dilated Convolutions and Gating
By Alice Coucke, Mohammed Chlieh, Thibault Gisselbrecht, David Leroy, Mathieu Poumeyrol, Thibaut Lavril. 2018
115 citations recorded by the source corpus.
We explore the application of end-to-end stateless temporal modeling to small-footprint keyword spotting as opposed to recurrent networks that model long-term temporal dependencies using internal states. We propose a model inspired by the recent success of dilated convolutions in sequence modeling applications, allowing to train deeper architectures in resource-constrained configurations. Gated activations and residual connections are also added, following a similar configuration to WaveNet. In addition, we apply a custom target labeling that back-propagates loss from specific frames of interest, therefore yielding higher accuracy and only requiring to detect the end of the keyword. Our experimental results show that our model outperforms a max-pooling loss trained recurrent neural network using LSTM cells, with a significant decrease in false rejection rate. The underlying dataset – "Hey Snips" utterances recorded by over 2.2K different speakers – has been made publicly available to establish an open reference for wake-word detection.
Subfields: Artificial Intelligence, Signal Processing
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Mirkó Visontai
- Oleg Rybakov
- Natasha Kononenko
- David Leroy
- Thibault Gisselbrecht
- Niranjan A. Subrahmanya
- Alice Coucke
- Thibaut Lavril
In Signal Processing
- Mirkó Visontai
- Oleg Rybakov
- Natasha Kononenko
- David Leroy
- Thibault Gisselbrecht
- Niranjan A. Subrahmanya
- Alice Coucke
- Thibaut Lavril
Citations in the corpus
- Federated Learning for Keyword Spotting (2018, methodology): “The use of of memory-efficient end-toend models [14] in place of the presented class-based model could ease labelling as it would only rely on voice activity detection.”
- Streaming keyword spotting on mobile devices (2020, methodology): “Another optimization approach is the transformation of the NN model (which was used during training) into another NN streaming model which can be more efficient for inference/prediction mode [4].”