AI and Advanced Analytics applied at
summary
Raiffeisen builds a proof-of-concept virtual call‑center agent for CEE languages that runs as a cloud prototype, ties voice‑to‑text to NER and dialogue logic, and stays mocked off core banking. Key components are Rasa for orchestration, spaCy (with custom NER) plus Facebook fastText embeddings via TensorFlow/Keras for multilingual intent, and SaaS voice TTS/STT with plans for Tacotron2. Prototype takes days with a small cross‑functional team, shows usable results for simple FX queries, exposes STT limits on Slavic languages, and raises GDPR, explainability and make‑vs‑buy tradeoffs.
AI-written summary · about
chapters (7)
- 0:00Start
- 8:22Customer Lifetime Value
- 19:00Virtual Call Center
- 23:19What Is the Lesson Learn
- 23:37Business Requirements and Goals Are Important
- 24:23Test Approach
- 31:48How Do You Manage the Identification Process and Prevent Identity Theft because Voices Are Easy To Fake
related talks
38mAI-Powered Conversational Interfaces
01h 03mAI Innovation for a Smarter Financial Ecosystem
57mAI has been achieved internally - The Journey
14mAI for a Billion Voices: Taking AI where there is Maximum Impact
52mAI Powered Contact Center Analytics
27mReinventing Customer Experience with Voice and Generative AI
26mPerspectives on Digital Transformation in Banking
23mContact Center Voice AI: Low-Latency Intelligence Extraction from Messy Audio Streams