Multi-Microphone Neural Speech Separation for Far-Field Multi-Talker Speech Recognition

By Takuya Yoshioka, Hakan Erdogan, Zhuo Chen, F. Alleva. 2018

121 citations recorded by the source corpus.

Subfields: Computational Mechanics, 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 Computational Mechanics

1. Jinyu Li
2. Sanyuan Chen
3. Shujie Liu
4. Yu Wu
5. Zhuo Chen
6. Naoyuki Kanda
7. Takuya Yoshioka
8. Dong Yu

In Signal Processing

1. Jinyu Li
2. Zhuo Chen
3. Dong Yu
4. Takuya Yoshioka
5. Shujie Liu
6. Yu Wu
7. Shinji Watanabe
8. Sanyuan Chen

Citations in the corpus

• WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing (2021, background): “Four different mixture types described in [73] are included in the training set.”
• Ultra Fast Speech Separation Model with Teacher Student Learning (2021, methodology): “Following [22], we include four different mixture types in the training data.”
• Continuous Speech Separation with Conformer (2020, methodology): “In this work, inspired by the recent advances in transducer-based end-to-end ASR modeling, which has evolved from a recurrent neu- ral network (RNN) transducer [17] to Transformer [18] and Conformer [19] transducers, we examine the use of the Conformer architecture for continuous speech separation (”
• Don’t Shoot Butterfly with Rifles: Multi-Channel Continuous Speech Separation with Early Exit Transformer (2020, methodology): “Following [24], we include four different mixture types in the training data.”
• Investigation of Practical Aspects of Single Channel Speech Separation for ASR (2021, background): “For two-speaker cases, the mixing SDR ranges from -5 dB to 5 dB and four mixture types are considered following the work [36].”
• Complex Neural Spatial Filter: Enhancing Multi-Channel Target Speech Separation in Complex Domain (2021, background): “Previous investigations on interaural phase difference (IPD) and its derived directional features showed that these features can provide discriminative cues for cRM estimation [1], [2], [16]–[19].”
• Continuous Speech Separation: Dataset and Analysis (2020, methodology): “In this paper, we investigate the speaker-independent CSS approach proposed in [18].”