Indian EmoSpeech Command Dataset: A dataset for emotion based speech recognition in the wild
This work addresses emotion recognition in speech for real-world applications like on-device models, but it is incremental as it focuses on a new dataset and modest performance improvements.
The authors tackled the problem of emotion analysis in speech by creating the Indian EmoSpeech Command Dataset, which includes both verbal commands and non-verbal background sounds, and achieved a 3.3% average gain in top-one score over a subset of a speech command dataset for keyword spotting.
Speech emotion analysis is an important task which further enables several application use cases. The non-verbal sounds within speech utterances also play a pivotal role in emotion analysis in speech. Due to the widespread use of smartphones, it becomes viable to analyze speech commands captured using microphones for emotion understanding by utilizing on-device machine learning models. The non-verbal information includes the environment background sounds describing the type of surroundings, current situation and activities being performed. In this work, we consider both verbal (speech commands) and non-verbal sounds (background noises) within an utterance for emotion analysis in real-life scenarios. We create an indigenous dataset for this task namely "Indian EmoSpeech Command Dataset". It contains keywords with diverse emotions and background sounds, presented to explore new challenges in audio analysis. We exhaustively compare with various baseline models for emotion analysis on speech commands on several performance metrics. We demonstrate that we achieve a significant average gain of 3.3% in top-one score over a subset of speech command dataset for keyword spotting.