5 papers
Experience-Calibrated Contrastive Decoding for Mitigating Hallucinations in LM-Based Text-to-Speech
Chenlin Liu, Minghui Fang, Zhonghao Bi +3
Language model-based text-to-speech (LM-based TTS) remains vulnerable to speech hallucinations that deviate from the target text. Existing mitigation mainly relies on architectural…
Knowledge-Decoupled Functionally Invariant Path with Synthetic Personal Data for Personalized ASR
Yue Gu, Zhihao Du, Ying Shi +2
Fine-tuning generic ASR models with large-scale synthetic personal data can enhance the personalization of ASR models, but it introduces challenges in adapting to synthetic persona…
Enhancing the Robustness of Contextual ASR to Varying Biasing Information Volumes Through Purified Semantic Correlation Joint Modeling
Yue Gu, Zhihao Du, Ying Shi +3
Recently, cross-attention-based contextual automatic speech recognition (ASR) models have made notable advancements in recognizing personalized biasing phrases. However, the effect…
Mitigating Hallucinations in LM-Based TTS Models via Distribution Alignment Using GFlowNets
Chenlin Liu, Minghui Fang, Patrick Zhang +3
Language Model (LM)-based Text-to-Speech (TTS) systems often generate hallucinated speech that deviates from input text. Existing mitigation strategies either demand excessive trai…
Serialized Output Training by Learned Dominance
Ying Shi, Lantian Li, Shi Yin +2
Serialized Output Training (SOT) has showcased state-of-the-art performance in multi-talker speech recognition by sequentially decoding the speech of individual speakers. To addres…