Object & scene recognition
Detects people, objects, vehicles, animals, text (OCR), layout, selection marks, barcodes and formulas in images and videos – with high accuracy.
Coming soon
showmi
Every image. Every video. Every recording. Structured knowledge.
showmi turns unstructured photo, video, audio and document data into validated, structured information – with confidence scores, source references and schema-based output for your business processes.
Features
What showmi can do
Multimodal: images, videos, audio and documents. Schema-based, with confidence scores and source grounding – GDPR-compliant on Azure in Germany.
Speech & speaker recognition
Transcription with speaker diarization for audio and video, including timestamps – for full traceability.
Indexing & RAG
Markdown and vector index of all recognized content – instantly searchable and ready for Retrieval-Augmented Generation.
Semantic search
Search in natural language for scenes, people or situations: "Show all videos in which product X is presented."
AI chatbot with source grounding
Ask questions about your photo, video and audio archives. showmi answers with a reference to frame, region or timestamp.
Ecosystem integration
showmi works seamlessly with tellmi (call recording), talkmi (translation), askmi (knowledge) and notemi (documents).
Schema-based field extraction
Define your own fields via JSON schema – extract, classify or generate. Up to 1,000 fields per analyzer.
Confidence & source grounding
Every extracted value comes with a confidence score (0–1) and a reference to its source – enabling reliable straight-through processing.
Classification & segmentation
Videos are automatically segmented into scenes and content is pre-classified – up to 300 categories per field.
Use cases
How companies use showmi
Quality assurance: classify defect images automatically
showmi classifies production photos (crack, scratch, discoloration) and returns a confidence score per finding – only edge cases need manual review.
Quality assuranceClassificationConfidence scores
Media archive: search videos scene by scene
Videos are segmented into scenes, speakers are identified and content is indexed. One click jumps to the right second in the video.
Media managementSegmentationDiarization
Site security & access documentation
Person and vehicle detection in camera recordings, with classification and timestamps – audit-proof documentation on Azure in Germany.
Site securityAccess logGDPR
Damage & insurance documentation
Photo plus voice note become a structured damage record: category, estimated value, description – ready for your claims system.
InsuranceClaimsAutomation
Post-call & meeting insights
Meeting audio and video become summaries, action items, sentiment and KPI fields – with timestamps as source references.
Meeting insightsSentimentKPIs
Platform & options
Built on Azure Content Understanding
showmi uses Microsoft's multimodal Foundry service (GA, API 2025-11-01). Four modalities in a single pipeline – hosted GDPR-compliant in Germany.
Supported modalities
- ▸Documents: PDF, Office, TXT/HTML – up to 200 MB / 300 pages
- ▸Images: JPG, PNG, HEIF, BMP, TIFF – up to 200 MB
- ▸Audio: MP3, WAV, M4A, FLAC, OGG – up to 1 GB / 4 h
- ▸Video: MP4, MOV, MKV, AVI – up to 4 GB / 2 h
Options
Pro Mode Additional cost
Higher accuracy for complex documents and images via extended model usage. Available as a paid optional add-on.
Configurable content filters Optional
Safety thresholds for hate, violence, sexual content and self-harm can be adjusted – useful for example for private end-users with their own requirements.
GDPR-compliant
100% EU-compliant
Azure Germany
Frankfurt & Berlin
Coming soon
Request early access
Azure Content Understanding
Microsoft Foundry
Cloud Deployment
100% SaaS – no installation
All products run entirely in the cloud. No download, no setup, no IT department needed – just open your browser and get started.
Compatible with all common operating systems:
iOSAndroidWindowsmacOSLinux
Ready to leverage your image, video and audio data?
showmi is coming soon. Sign up for early access now and help shape the product with us.
No spam. No sharing with third parties. GDPR-compliant.