No abstract

No abstract

Motivational interviewing is a goal-oriented psychotherapy, employed in cases such as addiction, that aims to help clients explore and resolve their ambivalence about their problem. In motivational interviewing, it is desirable for the counselor to communicate empathy towards the client to promote better therapy outcomes. In this paper, we propose a deep neural network (DNN) system for predicting counselors' session level empathy ratings from transcripts of the interactions. First, we train a recurrent neural network mapping the text of each speaker turn to a set of task-specific behavioral acts that represent local dynamics of the client-counselor interaction. Subsequently, this network is used to initialize lower layers of a deep network predicting session level counselor empathy rating. We show that this method outperforms training the DNN end-to-end in a single stage and also outperforms a baseline neural network model that attempts to predict empathy ratings directly from text without modeling turn level behavioral dynamics. Index Terms: behavioral signal processing, recurrent neural networks, word embedding, motivational interviews 1 Note: we italicize empathy to distinguish this specific operational definition from colloquial definitions.

No abstract

The spoken term detection (STD) task aims to return relevant segments from a spoken archive that contain the query terms whether or not they are in the system vocabulary. This paper focuses on pronunciation modeling for Out-of-Vocabulary (OOV) terms which frequently occur in STD queries. The STD system described in this paper indexes word-level and sub-word level lattices or confusion networks produced by an LVCSR system using Weighted Finite State Transducers (WFST). We investigate the inclusion of n-best pronunciation variants for OOV terms (obtained from letter-to-sound rules) into the search and present the results obtained by indexing confusion networks as well as lattices. The following observations are worth mentioning: phone indexes generated from sub-words represent OOVs well and too many variants for the OOV terms degrade performance if pronunciations are not weighted.

The dissemination and evaluation of evidence based behavioral treatments for substance abuse problems rely on the evaluation of counselor interventions. In Motivational Interviewing (MI), a treatment that directs the therapist to utilize a particular linguistic style, proficiency is assessed via behavioral coding - a time consuming, non-technological approach. Natural language processing techniques have the potential to scale up the evaluation of behavioral treatments like MI. We present a novel computational approach to assessing components of MI, focusing on one specific counselor behavior – reflections – that are believed to be a critical MI ingredient. Using 57 sessions from 3 MI clinical trials, we automatically detected counselor reflections in a Maximum Entropy Markov Modeling framework using the raw linguistic data derived from session transcripts. We achieved 93% recall, 90% specificity, and 73% precision. Results provide insight into the linguistic information used by coders to make ratings and demonstrate the feasibility of new computational approaches to scaling up the evaluation of behavioral treatments.

No abstract

In psychotherapy interactions there are several desirable and undesirable behaviors that give insight into the efficacy of the counselor and the progress of the client. It is important to be able to identify when these target behaviors occur and what aspects of the interaction signal their occurrence. Manual observation and annotation of these behaviors is costly and time intensive. In this paper, we use long short term memory networks equipped with an attention mechanism to process transcripts of addiction counseling sessions and predict prominent counselor and client behaviors. We demonstrate that this approach gives competitive performance while also providing additional interpretability.