BSS Transitions 'Speech Analytics' to Autonomous AI Agents

BSS has unveiled major update 2.15 of its 'Speech Analytics' product, with the key focus on evolving the embedded AI from a reactive request-processing tool into a full-fledged proactive AI agent capable of independently managing customer service quality. The main innovation is the transformation of the insight mode into an autonomous insight agent that, on a schedule, analyzes call collections, identifies patterns and anomalies, generates insights, and automatically sends reports to stakeholders, including related departments. The integration of RAG (Retrieval-Augmented Generation) into automatic evaluation cards enables semantic quality control based on corporate knowledge bases or uploaded regulations without complex integration. For quality control departments, the release improves the sampling mechanism, adds an interactive 'Marker Tree' widget, a new 'Score Distribution' report, and reference books for feedback and categorization. The agent-trainer now supports language selection, multi-level evaluation (including critical goals that instantly terminate simulations), and detailed justification for each goal. Learning results are aggregated in a new 'Student Dashboard'. The platform also underwent a large-scale UX/UI refactoring, including modern components with advanced filtering, an adaptive dialog statistics widget, improved report builder, and transcription of audio messages in text chats. Anna Ivleva, product owner of Speech Analytics at BSS, noted that the update blurs the line between an analytical tool and a digital employee, delivering a full-fledged AI agent that proactively monitors communications and delivers insights to businesses.