116 citations · 808 across the 43 of their papers we have counts for
17 papers · 1 filter
Minimum Word Error Rate Training with Language Model Fusion for End-to-End Speech Recognition
Zhong Meng, Yu Wu, Naoyuki Kanda +6
Integrating external language models (LMs) into end-to-end (E2E) models remains a challenging task for domain-adaptive speech recognition. Recently, internal language model estimat…
On Addressing Practical Challenges for RNN-Transducer
Rui Zhao, Jian Xue, Jinyu Li +3
In this paper, several works are proposed to address practical challenges for deploying RNN Transducer (RNN-T) based speech recognition system. These challenges are adapting a well…
Internal Language Model Training for Domain-Adaptive End-to-End Speech Recognition
Zhong Meng, Naoyuki Kanda, Yashesh Gaur +6
The efficacy of external language model (LM) integration with existing end-to-end (E2E) automatic speech recognition (ASR) systems can be improved significantly using the internal…
Internal Language Model Estimation for Domain-Adaptive End-to-End Speech Recognition
Zhong Meng, Sarangarajan Parthasarathy, Eric Sun +7
The external language models (LM) integration remains a challenging task for end-to-end (E2E) automatic speech recognition (ASR) which has no clear division between acoustic and la…
Microsoft Speaker Diarization System for the VoxCeleb Speaker Recognition Challenge 2020
Xiong Xiao, Naoyuki Kanda, Zhuo Chen +10
This paper describes the Microsoft speaker diarization system for monaural multi-talker recordings in the wild, evaluated at the diarization track of the VoxCeleb Speaker Recogniti…
Developing RNN-T Models Surpassing High-Performance Hybrid Models with Customization Capability
Jinyu Li, Rui Zhao, Zhong Meng +8
Because of its streaming nature, recurrent neural network transducer (RNN-T) is a very promising end-to-end (E2E) model that may replace the popular hybrid model for automatic spee…
Exploring Transformers for Large-Scale Speech Recognition
Liang Lu, Changliang Liu, Jinyu Li +1
While recurrent neural networks still largely define state-of-the-art speech recognition systems, the Transformer network has been proven to be a competitive alternative, especiall…
L-Vector: Neural Label Embedding for Domain Adaptation
Zhong Meng, Hu Hu, Jinyu Li +4
We propose a novel neural label embedding (NLE) scheme for the domain adaptation of a deep neural network (DNN) acoustic model with unpaired data samples from source and target dom…
High-Accuracy and Low-Latency Speech Recognition with Two-Head Contextual Layer Trajectory LSTM Model
Jinyu Li, Rui Zhao, Eric Sun +4
While the community keeps promoting end-to-end models over conventional hybrid models, which usually are long short-term memory (LSTM) models trained with a cross entropy criterion…
Domain Adaptation via Teacher-Student Learning for End-to-End Speech Recognition
Zhong Meng, Jinyu Li, Yashesh Gaur +1
Teacher-student (T/S) has shown to be effective for domain adaptation of deep neural network acoustic models in hybrid speech recognition systems. In this work, we extend the T/S l…
Character-Aware Attention-Based End-to-End Speech Recognition
Zhong Meng, Yashesh Gaur, Jinyu Li +1
Predicting words and subword units (WSUs) as the output has shown to be effective for the attention-based encoder-decoder (AED) model in end-to-end speech recognition. However, as…
Advances in Online Audio-Visual Meeting Transcription
Takuya Yoshioka, Igor Abramovski, Cem Aksoylar +23
This paper describes a system that generates speaker-annotated transcripts of meetings by using a microphone array and a 360-degree camera. The hallmark of the system is its abilit…
Self-Teaching Networks
Liang Lu, Eric Sun, Yifan Gong
We propose self-teaching networks to improve the generalization capacity of deep neural networks. The idea is to generate soft supervision labels using the output layer for trainin…
Cycle-Consistent Speech Enhancement
Zhong Meng, Jinyu Li, Yifan Gong +2
Feature mapping using deep neural networks is an effective approach for single-channel speech enhancement. Noisy features are transformed to the enhanced ones through a mapping net…
Adversarial Feature-Mapping for Speech Enhancement
Zhong Meng, Jinyu Li, Yifan Gong +2
Feature-mapping with deep neural networks is commonly used for single-channel speech enhancement, in which a feature-mapping network directly transforms the noisy features to the c…
Speaker-Invariant Training via Adversarial Learning
Zhong Meng, Jinyu Li, Zhuo Chen +5
We propose a novel adversarial multi-task learning scheme, aiming at actively curtailing the inter-talker feature variability while maximizing its senone discriminability so as to…
Adversarial Teacher-Student Learning for Unsupervised Domain Adaptation
Zhong Meng, Jinyu Li, Yifan Gong +2
The teacher-student (T/S) learning has been shown effective in unsupervised domain adaptation [1]. It is a form of transfer learning, not in terms of the transfer of recognition de…