Text Mining API
Trained meticulously on our proprietary datasets, our text mining solution delivers 30% better performance* on average over open-source alternatives, while ensuring full transparency and accessibility from day one.

See how Lettria's Sentiment Analysis outperforms other options
Read a detailed analysis of the performance of the world's most widely used models for sentiment analysis tasks. See how Lettria compares and outperforms the rest.

Amplify Insights with Sentiment Analysis
Leverage Lettria's Deep Learning-powered Sentiment Analysis, that is 30% more accurate* than a comparable open source model.
Our advanced solution deciphers and quantifies sentiment from -1 to 1 and detects different sentiments in different parts of the same sentence, allowing you to propel informed decisions, optimize brand reputation, and enhance customer experience with actionable insights.


Unleash Advanced Emotion Analysis
Empower your business with Lettria's deep learning-powered Emotion Analysis, capable of detecting up to 28 distinct emotions in text. A game-changer for elevating customer satisfaction, gauging public sentiment, and enriching user experiences.
Understand meaning with Entity Recognition
Lettria's Entity Recognition capabilities go way beyond standard open-source NER tools. In the same text, you can recognize regular entities, numeric entities, as well as meanings.
Identify a cappuccino as a "hot beverage" or a MacBook as an "electronic device", while extracting email IDs, phone numbers and scientific measurements from your text. The possibilities are endless.

%20(1).png)
Patrick Duvaut
Head of Innovation
%20(1).png)
Patrick Duvaut
Head of Innovation
%20(1).png)
Patrick Duvaut
Head of Innovation
Frequently Asked Questions
Lettria is ideal for organizations in finance, insurance, healthcare, legal, and life sciences—anywhere unstructured or semi‑structured documents (contracts, claims, reports) must be processed with structure, clarity, and auditability
The platform includes dashboards for usage metrics, extraction accuracy, workflow monitoring, and audit logs—supporting compliance tracking and process visibility .
Lettria’s GraphRAG engine preserves original document structures and traces each extracted element back to its source. This enables explainable, auditable outputs—essential in regulated domains
Yes—our studio offers drag‑and‑drop document uploads, built‑in ML tools, prompt engineering, and guided AutoML via AutoLettria. Business users can spin up prototypes in days (not months) and deploy production flows via APIs
The platform is API-first, with over 50 native connectors (e.g., SharePoint, S3) and Power Automate logic app support. Lettria embeds into existing workflows via APIs and webhooks.
Data is hosted on European servers with end-to-end encryption and SSL/TLS. We’re GDPR-compliant and undergo regular audits with ANSSI collaboration—ensuring enterprise-grade privacy and trust
Our Knowledge Studio includes dictionary, taxonomy, and ontology modules. You can enrich dictionaries with auto-detected keywords and upload/import taxonomies or ontologies (.rdf, .owl, .skos) for editorial and collaborative enrichment
Lettria delivers a full suite of NLP and AI tools: OCR ingestion, text cleaning, text labeling, table parsing, classification, entity recognition, semantic search, batch annotation, prompt engineering, and robust analytics dashboards—designed for complex document workflows