13 citations · 14 across the 3 of their papers we have counts for
5 papers
Preference Alignment Improves Language Model-Based TTS
Jinchuan Tian, Chunlei Zhang, Jiatong Shi +4
Recent advancements in text-to-speech (TTS) have shown that language model (LM)-based systems offer competitive performance to their counterparts. Further optimization can be achie…
Speaker-Aware Mixture of Mixtures Training for Weakly Supervised Speaker Extraction
Zifeng Zhao, Rongzhi Gu, Dongchao Yang +2
Dominant researches adopt supervised training for speaker extraction, while the scarcity of ideally clean corpus and channel mismatch problem are rarely considered. To this end, we…
Improving Mandarin End-to-End Speech Recognition with Word N-gram Language Model
Jinchuan Tian, Jianwei Yu, Chao Weng +2
Despite the rapid progress of end-to-end (E2E) automatic speech recognition (ASR), it has been shown that incorporating external language models (LMs) into the decoding can further…
Layer Reduction: Accelerating Conformer-Based Self-Supervised Model via Layer Consistency
Jinchuan Tian, Rongzhi Gu, Helin Wang +1
Transformer-based self-supervised models are trained as feature extractors and have empowered many downstream speech tasks to achieve state-of-the-art performance. However, both th…
A Random Gossip BMUF Process for Neural Language Modeling
Yiheng Huang, Jinchuan Tian, Lei Han +4
Neural network language model (NNLM) is an essential component of industrial ASR systems. One important challenge of training an NNLM is to leverage between scaling the learning pr…