Topic Identification For Spontaneous Speech: Enriching Audio Features With Embedded Linguistic Information
This work addresses the problem of degraded performance in topic identification for spontaneous speech in low-resource settings, offering incremental improvements over traditional text-based methods.
The paper tackled topic identification from spontaneous speech in low-resource scenarios by comparing audio-only and hybrid models, finding that hybrid multi-modal solutions achieved the best results while audio-only methods were viable when ASR was unavailable.
Traditional topic identification solutions from audio rely on an automatic speech recognition system (ASR) to produce transcripts used as input to a text-based model. These approaches work well in high-resource scenarios, where there are sufficient data to train both components of the pipeline. However, in low-resource situations, the ASR system, even if available, produces low-quality transcripts, leading to a bad text-based classifier. Moreover, spontaneous speech containing hesitations can further degrade the performance of the ASR model. In this paper, we investigate alternatives to the standard text-only solutions by comparing audio-only and hybrid techniques of jointly utilising text and audio features. The models evaluated on spontaneous Finnish speech demonstrate that purely audio-based solutions are a viable option when ASR components are not available, while the hybrid multi-modal solutions achieve the best results.