October 2024 Summaries

Genesys Cloud is a contact center solution used by organizations to manage customer interactions across multiple channels. By integrating Genesys Cloud with Symbl's conversational intelligence, companies can extract valuable insights from interactions and optimize their contact center performance. Symbl's Call Score API provides a numerical score for each conversation, along with an individual score and detailed breakdown for defined criteria. This allows organizations to evaluate the quality of conversations, assess agent performance, refine communication guidelines, ensure compliance, and improve recruitment processes. The integration process involves setting up Genesys Cloud, configuring AudioHook Monitor, customizing the Call Score request, selecting which interactions receive call scores, and activating the integration.

In this tutorial, you will learn how to build an AI call center assistant using Symbl's intelligence APIs. The solution involves streaming real-time audio from Amazon Connect to Symbl via Amazon Kinesis and utilizing Trackers, Nebula LLM, and retrieval augmented generation (RAG) to provide agents with real-time troubleshooting tips during phone conversations with customers. This AI call center assistant can improve agent performance by 34% and potentially offer even greater gains in the future.

The integration of Symbl.ai's contextual AI with Snowflake's data capabilities can help organizations turn conversations into powerful insights and actions. This partnership enables three core capabilities: Search, Data Enrichment, and Centralized Agentic Workflows. By making interaction data searchable, actionable, and a true source of business intelligence, businesses can improve their analytics, empower search-driven insights, and build agentic workflows that drive impactful business outcomes.

Symbl.ai introduces its Real-Time Assist API, a new addition to their suite of generative APIs designed for enterprises. The API enables advanced real-time AI capabilities such as contextual guidance for handling objections, compliance violations, and script adherence without requiring extensive development or operational overhead. This solution accelerates time-to-value for live assistance from months to days while maintaining quality. It addresses the challenges faced by enterprises in building effective real-time agent assistance capabilities, including fragmented technology stacks, delayed time-to-market, and maintenance overhead. The Real-Time Assist API offers a unified platform for real-time guidance, accelerated time-to-value, and minimal maintenance with continuous improvement. Key features include objection handling, script adherence, compliance monitoring, and real-time Q&A.

Streaming databases are essential for building real-time Generative AI (GenAI) applications due to their ability to process continuous data streams in real time, enabling immediate actions and insights within applications and systems. They support a wide range of real-time applications across all industries by collecting, processing, and analyzing data as soon as it's available. Key benefits for GenAI applications include real-time data processing, scalability and performance, and easy integration with AI tools. Top streaming databases include open-source options like Apache Kafka, RisingWave, and Arroyo; source-available platforms such as KsqlDB, Materialize, and EventStoreDB; and closed-source solutions like Timeplus and DeltaStream. When selecting a streaming database for GenAI applications, consider performance metrics, scalability, ease of use and integration, and cost considerations.

Large Language Models (LLMs) have transformed how organizations approach their work, offering capabilities such as question-answering, sentiment analysis, text synthesis, and summarization. Companies can choose between open-source or closed-source LLMs for their projects. Open-source LLMs are characterized by their publicly available source code, allowing anyone to use, modify, and distribute them. They offer control, transparency, community support, and cost-effectiveness but may be less secure and stable than closed-source models. Closed-source LLMs are proprietary models developed and maintained by private vendors with a dedicated team of experts. They provide more performant and robust models, proprietary innovations, simpler integration, consistency, and support but come at a cost and raise privacy concerns. The choice between open-source or closed-source LLMs depends on an organization's specific needs and available resources.