Innovations

Virtual Lab chatbot/voicebot

Problems addressed

Non-expert users and interdisciplinary researchers face steep learning curves and slow manual navigation when searching, querying, and attempting to utilize complex technical datasets and documentation.

Unique Value Proposition

Culturally aware Natural Language Processing (NLP) text and voice assistant integrated into the Virtual Lab platform, enabling plain-language interactions to query cross-domain data spaces and Q&A knowledge bases.

Validated UC & Key Performance Benchmark

(Where was it tested, and what exact performance improvement or resource reduction did it achieve?)

Tested in: Integrated with AETHON’s Virtual Lab platform for cross-domain dataset Q&A interaction.

Hard metrics:
30+ languages supported with automatic detection (no configuration required)
RSA-2048 per-token encryption on the VirtualLab channel
10-turn conversation history maintained per session
Top-3 document retrieval via pgvector cosine similarity (threshold 0.55, 384-dim embeddings)
Voice input supported via Whisper STT (WebM audio → transcription → LLM response)
2 integrated platforms: VirtualLab (live), NCSRD (in progress)

Target Stakeholder Segments

(Who adopts/Hands-on Users)

Researchers, Data Scientists, Municipal Staff, Students, University Deans, and Regional Government Innovation Officers

Key Industry / Vertical Markets

(Where it is deployed)

Universities, Research Organisations, Local Municipalities, and European Digital Innovation Hubs (EDIHs)

License Model & IP and Commercial Strategy

(Open Source License)

Open Source License:
Current licenses of VL: MIT License, BSD-3-Clause, CC0-1.0, GPL
Current licenses of USAL’s chatbot component: Apache 2.0

Commercial Streams: Virtual Lab SaaS subscriptions; Custom integration SLAs;

Proprietary Strategy: Software & Content Copyright

Availability

(Code Repository link)

Open Source License:
Current licenses of VL: MIT License, BSD-3-Clause, CC0-1.0, GPL
Current licenses of USAL’s chatbot component: Apache 2.0

Commercial Streams: Virtual Lab SaaS subscriptions; Custom integration SLAs;

Proprietary Strategy: Software & Content Copyright

Estimated Time to Market

1–2 years post-project