alpha-avatar-plugins-persona 0.5.4

AlphaAvatar Framework plugin for persona service

Persona Plugin for AlphaAvatar

Support for automatic extraction and real-time matching of user full-modality personas, enabling Avatar to recognize, track, and personalize interactions based on multimodal signals. This plugin provides a unified interface for profiling (semantic + vector representation), speaker identification, and face recognition, abstracting away backend complexity so developers can focus on building adaptive, personalized experiences.

Features

• Automatic Persona Extraction: On-the-fly profiling of users based on conversation history, behavioral cues, and multimodal inputs. Persona profiles are represented as vector embeddings for efficient similarity search and personalization.

• Real-time Persona Matching: Incoming dialogue and multimodal signals are matched against stored persona profiles in real-time, ensuring the Avatar can recall context, traits, and preferences instantly.

Installation

pip install alpha-avatar-plugins-persona

---

Supported Modules

Vector Store

Module	Description	Docs
Qdrant (default)	Combined framework for persona extraction pipelines and vector storage / retrieval.	Qdrant

Profile

Module	Description	Docs
LangChain (default)	Combined framework for persona extraction pipelines and vector storage / retrieval.	LangChain

Speaker

Module	Description	Docs
ERes2NetV2 (default)	State-of-the-art deep speaker recognition model for speaker embedding and identification.	3D-Speaker
wav2vec2-large-robust-6-ft-age-gender (default)	Deep learning model for speaker embedding with additional age and gender recognition capabilities.	audEERING

Face Recognizer

Module	Description	Docs
Face ID (default)	Face embedding + recognition module for identity verification and multimodal persona binding.	(e.g., InsightFace)