Speech Prefix-Tuning with RNNT Loss for Improving LLM Predictions
In this paper, we focus on addressing the constraints faced when applying LLMs to ASR.
Recent works utilize prefixLM-type models, which directly apply speech as a prefix to LLMs …
Recent works utilize prefixLM-type models, which directly apply speech as a prefix to LLMs …
Tuning Large Language Model for Speech Recognition With Mixed-Scale Re-Tokenization
Large Language Models (LLMs) have proven successful across a spectrum of speech-
related tasks, such as speech recognition, text-to-speech, and spoken language …
related tasks, such as speech recognition, text-to-speech, and spoken language …
MaLa-ASR: Multimedia-Assisted LLM-Based ASR
As more and more information-rich data like video become available, utilizing multi-modal
auxiliary information to enhance audio tasks has sparked widespread research interest. The …
auxiliary information to enhance audio tasks has sparked widespread research interest. The …
Pronunciation Assessment with Multi-modal Large Language Models
Large language models (LLMs), renowned for their powerful conversational abilities, are
widely recognized as exceptional tools in the field of education, particularly in the context of …
widely recognized as exceptional tools in the field of education, particularly in the context of …
Unveiling the Potential of LLM-Based ASR on Chinese Open-Source Datasets
Large Language Models have demonstrated unparalleled effectiveness in various NLP
tasks, and integrating LLMs with automatic speech recognition is becoming a mainstream …
tasks, and integrating LLMs with automatic speech recognition is becoming a mainstream …
MooER: LLM-based Speech Recognition and Translation Models from Moore Threads
J Xu, Z Liang, Y Liu, Y Hu, J Li, Y Zheng, M Cai… - arXiv preprint arXiv …, 2024 - arxiv.org
In this paper, we present MooER, a LLM-based large-scale automatic speech recognition
(ASR)/automatic speech translation (AST) model of Moore Threads. A 5000h pseudo …
(ASR)/automatic speech translation (AST) model of Moore Threads. A 5000h pseudo …
StreamVoice+: Evolving into End-to-end Streaming Zero-shot Voice Conversion
StreamVoice has recently pushed the boundaries of zero-shot voice conversion (VC) in the
streaming domain. It uses a streamable language model (LM) with a context-aware …
streaming domain. It uses a streamable language model (LM) with a context-aware …