Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
97 tokens/sec
GPT-4o
53 tokens/sec
Gemini 2.5 Pro Pro
44 tokens/sec
o3 Pro
5 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Transfer learning of language-independent end-to-end ASR with language model fusion (1811.02134v2)

Published 6 Nov 2018 in cs.CL

Abstract: This work explores better adaptation methods to low-resource languages using an external LLM (LM) under the framework of transfer learning. We first build a language-independent ASR system in a unified sequence-to-sequence (S2S) architecture with a shared vocabulary among all languages. During adaptation, we perform LM fusion transfer, where an external LM is integrated into the decoder network of the attention-based S2S model in the whole adaptation stage, to effectively incorporate linguistic context of the target language. We also investigate various seed models for transfer learning. Experimental evaluations using the IARPA BABEL data set show that LM fusion transfer improves performances on all target five languages compared with simple transfer learning when the external text data is available. Our final system drastically reduces the performance gap from the hybrid systems.

User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (5)
  1. Hirofumi Inaguma (42 papers)
  2. Jaejin Cho (24 papers)
  3. Murali Karthick Baskar (15 papers)
  4. Tatsuya Kawahara (61 papers)
  5. Shinji Watanabe (416 papers)
Citations (39)

Summary

We haven't generated a summary for this paper yet.