Abstract: We study speech emotion recognition based on linguistic features that consider the spoken language in Japanese. In this approach, speech recognition is used to convert speech into text. The ...
Abstract: The latency and computation demand of End-to-end (E2E) automatic speech recognition (ASR) models hinder their deployment on lightweight devices. Despite there are many methods proposed for ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback