The speech recognition specialist Deepgram has received $12 million in Series A funding. NVIDIA, Y Combinator, Compound, and SAP.iO, all participated in the funding round, though the round was ultimately led by Wing VC.

Deepgram plans to use the funds to build better speech recognition solutions for enterprise-level customers. The company is specifically working on technology that will allow companies to transcribe and analyze phone calls, meetings, and verbal communications in real time, thereby turning those conversations into actionable data that can be used for a variety of purposes.

According to Deepgram, its solution is unique because it utilizes end-to-end deep learning instead of heuristics-based speech processing. That makes the platform more efficient (and more scalable) than other speech recognition solutions.

“Deepgram is doing groundbreaking work in this field, and we are delighted to be working closely with them,” said NVIDIA Business Development VP Jeff Herbst. “Their GPU-accelerated speech recognition enables faster, more accurate natural language processing that will make an important impact on a range of industries.”

Deepgram went on to introduce two new features to commemorate its Series A success. The first is a streaming feature that transcribes speech in real time, which can cover everything from complex call center interactions to simple requests to a takeout order from a restaurant.

The second feature is On-Premises Deployment, which is designed for companies that need to transcribe confidential or sensitive information in a secure environment. The solution will ensure that the conversations are kept in-house.

Deepgram will be hoping to capture a portion of a lucrative speech recognition market as its technology continues to improve. Meticulous Research recently predicted that the demand for voice and speech recognition tech will reach $26.8 billion by 2025, while Grand View believes that the market will climb to nearly $32 billion in the same forecast period.

(Originally posted on FindBiometrics)