Towards Low-Resource Real-Time Assessment of Empathy in Counselling

Zixiu Wu, Rim Helaoui, Diego Reforgiato Recupero, Daniele Riboni · 2021

Gauging therapist empathy in counselling is an important component of understanding counselling quality.While session-level empathy assessment based on machine learning has been investigated extensively, it relies on relatively large amounts of well-annotated dialogue data, and real-time evaluation has been overlooked in the past.In this paper, we focus on the task of low-resource utterance-level binary empathy assessment.We train deep learning models on heuristically constructed empathy vs. non-empathy contrast in general conversations, and apply the models directly to therapeutic dialogues, assuming correlation between empathy manifested in those two domains.We show that such training yields poor performance in general, probe its causes, and examine the actual effect of learning from empathy contrast in general conversation.r/offmychest ... Speaker: being married to a depressed person is so lonely .that is all .thanks for listening .Listener: sorry to say this but it 's not worth being in a relationship if both of you are n't happy .r/CasualConversation ... Speaker: can you recommend me some good music ?i 've decided to expand my taste in music and i need some advice .what are your favourite songs , redditors ?Listener: something corporate , jimmy eat world , and fall out boy are my top three recommendations

Read the paper · More papers on PaperTik