| Autor: |
Akihiko Nakagawa, Ernesto Damiani, Eriko Sakurai, Rainer Knauf, Andrea Kutics, Yukino Ikegami, Yoshitaka Sakurai, Setsuo Tsuruta |
| Rok vydání: |
2019 |
| Předmět: |
|
| Zdroj: |
SITIS |
| DOI: |
10.1109/sitis.2019.00081 |
| Popis: |
This paper introduces a speech recognition framework for high performance personalized adaption. It is based on plural language models and personalized incremental learning interface for error correction. If an error in a recognition result is detected by a bidirectional neural language model, it generates a corrected sentence by a majority decision among multiple n-gram language models considering several aspects. Moreover, we introduce a speaker adaptation by updating language models through incremental learning, which can adjust the parameter from training data. The experiments show that our framework improves word-error rate to 78% compared with Google Chrome's Speech Recognition API. Our framework can be used for improving one-to-one human-machine dialogue systems such as intelligent (counseling) agents. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
