SkillSync AI - Real-time matching system for service requests with dynamic priority and target price control

Abstract

System for the automated acceptance and structured processing of service requests via multiple input channels (text, voice, telephone), with an AI-based component analyzing natural language and extracting structured data.

Description

1. Technical field

[0001] The system is an AI-powered platform for the structured receipt, analysis, and referral of service requests. SkillSync AI enables the generation, analysis, validation, and dynamic processing of customer requests across multiple input channels (text, voice, telephone) with real-time processing through artificial intelligence. The goal is the intelligent generation of a service contract based on structured user data and market prices. 2. Main Technical Components 2.1 Input Processor

[0002] Receives customer inquiries via text, voice, or telephone. Uses NLP (e.g., BERT, SpaCy), STT (e.g., Whisper), and real-time dialogue systems (LLM + TTS + ASR) to extract structured information such as service type, location, and additional details. 2.2 Target Price Matcher

[0003] Compares the customer's specified target price with historical market data and price quotes from actual suppliers. Utilizes regression, vector analysis, and a trained service provider database. Importance of the target price:

[0004] The target price entered by the user is a key feature of the system. It serves not merely as a simple reference point, but forms the basis for the intelligent matching mechanism. By comparing the target price with dynamically calculated price ranges based on historical market data and supplier information, including the tradespeople's prices and service provider data registered in the system, the system ensures that the generated results directly correspond to the user's individual expectations and are not simply based on market averages. This feature gives the system a unique ability to personalize its services and significantly distinguishes it from conventional systems.

[0005] The target price can be specified by the user based on their own prior knowledge, an estimate, or a personal budget. The system uses this value as the starting point for the matching process.

[0006] To bring the results closer to the specified target price, the system calculates dynamic price ranges using statistical methods such as regression analysis and vector space models. These ranges are then compared with real-world supplier data, including tradespeople's prices registered in the system and other service provider information. Based on this, the system adjusts the selection of results so that the suggested services are as close as possible to the user's target price, while still respecting the user's filter and priority criteria. 2.3 Context Matching Unit (Context Matcher)

[0007] Uses decision trees, KNN and geolocation to assess the consistency of service, location, time, duration and availability. 2.4 Priority Filter

[0008] Allows weighting and filtering based on criteria such as speed, price, quality, safety, insurance, and warranties. Uses weighted scoring and rule-based logic. 2.5 Matching Output Unit

[0009] It generates the optimal result based on user priorities and target price. It supports dynamic changes and immediate response to user adjustments. 2.6 Voice Interaction Module

[0010] Controls complete speech interaction using ASR, LLM-based dialogue guidance, and TTS for natural responses. Recognizes and processes speech commands in real time. 2.7 User Interface (Dynamic UI)

[0011] Used for user interaction. Allows adjustment of filters, priorities, and prices, as well as real-time order confirmation. 2.8 Backend & Security Structure (Core Engine)

[0012] Consists of a Python / FastAPI-based backend with a PostgreSQL database, exclusive service provider data, data encryption and GDPR-compliant architecture. 3. Technical procedure 1.3.1 Request Input: The user enters a request via text, voice, or telephone. The system processes the content via (2.1) and structures it. 2.3.2 Target price check: Unit (2.2) compares the specified target price with the service provider database. 3.3 Context-based matching: (2.3) analyzes location, time, service type and other relevant parameters. 4. 3.4 Priority-based filtering: (2.4) filters providers according to the criteria set by the user. 5.3.5 Result generation: (2.5) displays the best possible result; the user can make changes. 6. 3.6 Voice interaction: (2.6) enables full control via voice (e.g. change price, confirm request). 7. 3.7 Order creation and confirmation: User confirms the proposal via the dynamic user interface (2.7). 8.3.8 Storage and execution: The system submits the final request for execution; the backend (2.8) documents the process. Notice of registered trademark:

[0013] This system is fully integrated into the application "Minestlimit," a trademark registered with the German Patent and Trademark Office and owned by the applicant, Alexandros Herz. The intellectual property right relates exclusively to the technical functionality of the "SkillSync AI" system, which operates under this trademark name. Technical structure & special features: • Backend: Python, FastAPI, PostgreSQL • AI: BERT, GPT, Whisper • UI: Dynamic interface, real-time response • Security: GDPR compliant, data encryption • Database: Exclusively trained with provider data SkillSync Al is the first system to combine the following features:

[0014] Multi-channel input, real-time voice interaction, automated order creation, target price matching with market data, and an industry-specific trained database.

[0015] The technical system “SkillSync AI” is fully integrated into the application “Minestlimit”, a trademark registered with the German Patent and Trademark Office (applicant: Alexandros Herz).

Claims

[1] System for the automated acceptance and structured processing of service requests via multiple input channels (text, speech, telephone), wherein an AI-based component analyzes natural language and extracts structured data. [2] System according to claim 1, characterized by , that a matching algorithm component is implemented which matches target prices with price ranges calculated by regression models and vector-based similarity analyses, with service provider data being updated in real time. [3] System according to any of the preceding claims, comprising a priority filtering unit which weights and adjusts results in real time according to user criteria (price, time, quality, safety, warranty, insurance). [4] System according to one of the preceding claims, wherein a speech unit uses ASR, TTS and LLM to enable dialog-driven process control with voice commands such as "change price" or "stop request". [5] System according to one of the preceding claims, implemented in a software architecture with FastAPI and PostgreSQL, equipped with a dynamic user interface for real-time control of filters, decisions and order confirmation. [6] System according to one of the preceding claims, which operates in compliance with the GDPR, uses encrypted data processing and documents all interactions.

Images