Natural language understanding is a form of natural language processing that enables machine reading comprehension. Common applications of this include AI assistants, chatbots, and voice assistants.
“Behind every chatbot and voice assistant lies a common piece of technology: Natural Language Understanding (NLU). Anytime a user interacts with an AI using natural language, their words need to be translated into a machine-readable description of what they meant. Voice requires an additional step, which is to transcribe the voice of the user into the corresponding text, prior to running the NLU,” Joseph Dureau, CTO of Snips, wrote in a post.
While the technology has advanced, it often requires a cloud service such as Microsoft, Google or Amazon. One company is trying to provide an alternative approach. Snips is a decentralized voice assistant technology and solution provider. The company announced it was open sourcing Snips NLU in March. Snips NLU enables natural language understanding that can run on the edge or on a server, with minimal footprint.
“[These] handful of providers made cloud-based NLU a commodity, powering all chatbots and voice assistants on the planet. Our solution runs on-device or on-premise with same or superior performances, minimal footprint, while being faster than a round trip to the cloud,” the team wrote in an email to SD Times “We made the decision to fully open our technology to capture a segment of this immense industry, and reduce the dependency of AI assistants worldwide to these centralized providers.”
Snips NLU is a Python library that can train models and use those trained models to make predictions about queries. The team has also open sourced Snips NLU-rs, a Rust implementation of the solution.
According to the team, the solution is able to provide the same or better speed, performance and accuracy as cloud providers.
The company also provides its voice platform for free, launched a web interface for creating custom assistants, and provides a data generation service for building conversational interfaces.
“Snips NLU is a fast, light, accurate and mature alternative to cloud-based Natural Language Understanding solutions. Open sourcing the code will further help us improve our technology, promote Privacy by Design principles, and bring AI to the edge,” Dureau wrote.