Natural Language Processing
Amigo is the most comprehensive CX & domain NLP engine, powered by a knowledge graphontology with pre-defined products, attributes, slang, abbreviations and language packs.
Product – Attribute - Product mapping for human like context specific conversations with personalized recommendations.
Polyglot Language Model
Understand mixed- language, slang etc. in the same model across languages.
Superior sentiment detection with subjectivity analysis, intensifiers & negation handling.
Context driven Multi-level intent analysis and joint slot filling.
We have amazing partners
Advanced Machine vision models trained and optimized for domain specific customer journeys -ID OCR, photo extraction, face verification, liveness detection.
Matching live streaming video with photo & other details extracted from IDs.
70+ parameters to detect liveness accurately.
Live video chat for authentication, Onboarding & sentiment..
Text extraction from ID’s and documents for authentication and processing
Affinsys speech engine is phonetically trained specifically for banking speech / voice recognition across languages, dialects, slang and abbreviations for highest accuracy.
Voice/speech verification for MFA and frictionless transactions.
Multi lingual ASR
ASR with language models for lower WER and higher accuracy.
Realtime emotion detection and analysis for tailored actions.
Conforming to standards like SSML, Voice XML , SISR, SRGS, WebRTC.
Amigo has one of the most advanced recommendation model with Ensemble Learning that combines multiple algorithms to deliver personalised recommendations for accurate targeting. Cognitive CX model fuses dynamic persona generation (based on Behavioural data analytics and past interactions) with context & Reinforcement learning for continuous improvement.
Dynamic Persona Generation
Industry specific models trained on your product and customer attributes
Identify closest persona in the customer journey
Dynamic persona matching with current context for higher accuracy
Live active stream of recommendations with reinforcement learning