Transcription with Scribe

Scribe uses artificial intelligence to extract high-quality data from telephony streams and enable voice workflows

AI for Financial Voice Streams


How it works

TRANSCRIPTION FOR THE FINANCIAL MARKETS

Scribe combines deep learning speech recognition with tailored financial language models to obtain the best accuracy for complex, noisy financial conversations.


Scribe Accuracy



AI THAT UNDERSTANDS FINANCE

Scribe automatically extracts quotes and trades from spoken conversations through our patented natural language processing libraries called Product Interpreters.


Scribe Interpreters



POWERFUL DATA EXTRACTION

Scribe’s Insight Engine summarizes conversations by extracting powerful insights, such as key terms, key phrases, speaker identities, and sentiment.


Scribe Insights



COMPLETE CUSTOMIZATION

Scribe’s API allows for end-to-end customization of acoustic and language models. Teach Scribe new words in a matter of minutes and new voices in a matter of hours.


Use Cases

STREAM TRANSCRIPTION

Scribe’s Stream Transcription engine creates time-stamped, searchable, speaker separated transcripts from voice streams or even audio files. Scribe’s Engine can be used via our public API or in an on-premise environment through our Docker container image. Our product interpreters and insight engine integrate directly. For cloud-friendly customers, Scribe can leverage alternative engines to obtain the most accurate transcript possible.


REAL-TIME QUOTE / TRADE / ORDER CAPTURE

Scribe’s Real-Time Capture engine listens to audio streams in real-time to extract quote, trade, or order metadata. Leverage Scribe’s data to populate a trade ticket or capture for general compliance purposes (e.g. MiFID II). Feed the engine audio files, real time streams from a computer via websocket, or dial a turret into to our call bot.



Interested in Scribe? Contact Us to get access.