Extracting Domain Invariant Features by Unsupervised Learning for Robust Automatic Speech Recognition

By Wei-Ning Hsu, James R. Glass. 2018

45 citations recorded by the source corpus.

The performance of automatic speech recognition (ASR) systems can be significantly compromised by previously unseen conditions, which is typically due to a mismatch between training and testing distributions. In this paper, we address robustness by studying domain invariant features, such that domain information becomes transparent to ASR systems, resolving the mismatch problem. Specifically, we investigate a recent model, called the Factorized Hierarchical Variational Autoencoder (FHVAE). FHVAEs learn to factorize sequence-level and segment-level attributes into different latent variables without supervision. We argue that the set of latent variables that contain segment -level information is our desired domain invariant feature for ASR. Experiments are conducted on Aurora-4 and CHiME-4, which demonstrate 41 % and 27% absolute word error rate reductions respectively on mismatched domains.

Subfields: Artificial Intelligence, Representation & Self-Supervised Learning, Speech & Audio, 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

In Representation & Self-Supervised Learning

In Speech & Audio

In Signal Processing

Citations in the corpus

Robust wav2vec 2.0: Analyzing Domain Shift in Self-Supervised Pre-Training (2021, background): “The first one is feature-based, which focuses on creating robust features [9, 10].”
Pushing the Limits of Semi-Supervised Learning for Automatic Speech Recognition (2020, background): “These networks are in turn fine-tuned on supervised data [37–43].”
Improved Noisy Student Training for Automatic Speech Recognition (2020, background): “In this approach, a separate task/loss to (pre-)train the acoustic model (or a part thereof) is introduced to help the model learn a good representation of the input, e.g., [24, 25, 26, 27, 28, 29, 30].”
Deep Contextualized Acoustic Representations for Semi-Supervised Speech Recognition (2019, background): “More recently, acoustic representation learning has drawn increasing attention [19, 20, 21, 22, 23, 24] in speech processing.”
Semi-Supervised Speech Recognition via Local Prior Matching (2020, methodology): “Unlike the proposed method, previous studies focus on learning domain invariant features (Sun et al., 2017; Hsu & Glass, 2018; Meng et al., 2017; 2018; 2019) or data augmentation through learned speech transformations (Hsu et al., 2017; 2018).”
A Study of Enhancement, Augmentation, and Autoencoder Methods for Domain Adaptation in Distant Speech Recognition (2018, background): “Following [40], since the generated feature sequence is 19 frame shorter, we repeat the first and the last feature vector at each end to match the original length.”
Scalable Factorized Hierarchical Variational Autoencoder Training (2018, background): “The ability to disentangle latent factors has been beneficial to a wide range of tasks, including domain adaptation [6], conditional data augmentation [7], and voice conversion [5].”
Transfer Learning from Audio-Visual Grounding to Speech Recognition (2019, methodology): “, Aurora-4 with all four conditions) [17, 8], we also evaluate FHVAE models trained on PlacesAudCap to test cross-dataset transferability, and on the subset of domains used for ASR training.”