Novel neural network based fusion for multistream ASR

By Sri Harish Reddy Mallidi, Hynek Hermansky. 2016

32 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Multimodal 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

1. Yu Zhang
2. Quoc V. Le
3. Yonghui Wu
4. Chung-Cheng Chiu
5. Bo Li
6. Jinyu Li
7. Tara N. Sainath
8. Shinji Watanabe

In Multimodal Learning

1. Bo Li
2. William Chan
3. Chung-Cheng Chiu
4. Daniel S. Park
5. Yu Zhang
6. Quoc V. Le
7. Yonghui Wu

In Speech & Audio

1. Yu Zhang
2. Yonghui Wu
3. Chung-Cheng Chiu
4. Quoc V. Le
5. Bo Li
6. Jinyu Li
7. William Chan
8. Daniel S. Park

In Signal Processing

1. Yu Zhang
2. Yonghui Wu
3. Chung-Cheng Chiu
4. Jinyu Li
5. Quoc V. Le
6. Bo Li
7. William Chan
8. Tara N. Sainath

Citations in the corpus

• SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition (2019, methodology): “Feature drop-outs have been employed for training multi-stream ASR systems [17].”
• Specaugment on Large Scale Datasets (2019, methodology): “Drop-out for features have been used for training multi-stream ASR systems in [14, 15].”
• Continual Learning in Automatic Speech Recognition (2020, unknown): “Multi-stream systems[23, 24, 25] combine the outputs of different neural models using a stream attention module which determines the relative confidence of each stream.”