Slang Labs launches its own Speech Recognition Engine

We launched our very own in-house Speech Recognition Engine to give users the Voice experiences they deserve.

Slang Labs is a Voice-Assistant-as-a-Service (VAaaS) company, now backed by Google. We provide a multilingual In-App Voice Assistant for mobile and web apps. The Voice Assistant can be easily integrated into existing mobile and web applications from various domains like retail e-commerce, travel, BFSI, Job and other domains, as it comes with pre-trained capabilities and knowledge about these domains

In our journey to integrate a Voice Assistant in every app, we’ve had to do many things; our latest advancement? The creation and launch of our very own in-house Speech Recognition Engine.

Trained on thousands of hours of English language data, our Speech Recognition Engine recognizes various accents and dialects, and has an accuracy of over 87%. It is built on top of a Deep Learning based model, meaning its performance is only expected to improve over time.

Satish Gupta, co-founder of Slang Labs, has been heading the efforts of building the ASR engine has this to say, "We want to provide the best voice experience for the Next Billion Users coming online, and having a high accuracy is critical. Slang ASR is our next generation Speech Recognition Engine, which will help us achieve our goal of reaching a billion Voice users."

Slang ASR engine boasts the following features, all crafted to provide the Next Billion users with the best Voice experience possible -

  1. Extensible: The technical architecture of Slang ASR allows the addition of new languages quickly and easily. New languages can be added with the support of a wide variety of accents and dialects.
  2. Training and Deployment: The ASR engine is cloud-based and can be dynamically hosted at different locations to provide sub-second latencies.
  3. Deep Learning Architecture: The Slang ASR is built on top of cutting-edge end-to-end Deep-Learning based models, that enables continuous improvement of the models for improved accuracy.

What are the next steps we are taking now that we have our own in-house speech recognition engine?

After extensively testing the Slang ASR Engine and confirming with a high degree of  confidence that it is better than the existing third party speech recognition service that we use today, we will start rolling it out to more customers. We are hopeful that this will provide even better accuracy that our customers have come to expect from Slang.

At Slang Labs we are committed to our vision of a Voice Assistant in every app. To learn more about us and other steps we’ve taken to make this a reality click here.

To have a more comprehensive conversation book a demo with us today!