SALM: Speech-augmented Language Model with In-context Learning for Speech Recognition and Translation
Chen, Zhehuai, Huang, He, Andrusenko, Andrei, Hrinchuk, Oleksii, Puvvada, Krishna C., Li, Jason, Ghosh, Subhankar, Balam, Jagadeesh, Ginsburg, Boris
–arXiv.org Artificial Intelligence
We present a novel Speech Augmented Language Model (SALM) with {\em multitask} and {\em in-context} learning capabilities. SALM comprises a frozen text LLM, a audio encoder, a modality adapter module, and LoRA layers to accommodate speech input and associated task instructions. The unified SALM not only achieves performance on par with task-specific Conformer baselines for Automatic Speech Recognition (ASR) and Speech Translation (AST), but also exhibits zero-shot in-context learning capabilities, demonstrated through keyword-boosting task for ASR and AST. Moreover, {\em speech supervised in-context training} is proposed to bridge the gap between LLM training and downstream speech tasks, which further boosts the in-context learning ability of speech-to-text models. Proposed model is open-sourced via NeMo toolkit.
arXiv.org Artificial Intelligence
Oct-13-2023