Speech ReaLLM -- Real-time Streaming Speech Recognition with Multimodal LLMs by Teaching the Flow of Time

Seide, Frank, Doulaty, Morrie, Shi, Yangyang, Gaur, Yashesh, Jia, Junteng, Wu, Chunyang

arXiv.org Artificial Intelligence 

We introduce Speech ReaLLM, a new ASR architecture that marries "decoder-only" ASR with the RNN-T to make multimodal LLM architectures capable of real-time streaming. This is the first "decoder-only" ASR architecture designed to handle continuous audio without explicit end-pointing. Speech ReaLLM is a special case of the more general ReaLLM ("real-time LLM") approach, also introduced here for the first time. The idea is inspired by RNN-T: Instead of generating a response only at the end of a user prompt, generate after every input token received in real time (it is often empty). On Librispeech "test", an 80M Speech ReaLLM achieves WERs of 3.0% and 7.4% in real time (without an external LM or auxiliary loss). This is only slightly above a 3x larger Attention-Encoder-Decoder baseline. We also show that this way, an LLM architecture can learn to represent and reproduce the flow of time; and that a pre-trained 7B LLM can be fine-tuned to do reasonably well on this task.

Duplicate Docs Excel Report

Title
None found

Similar Docs  Excel Report  more

TitleSimilaritySource
None found