Google today said developers will soon be able to add speech recognition to their apps through the Google Could Speech API.
Google is offering the same neural network-based tool it uses in its own voice search and keyboard apps. The API recognizes more than 80 languages and dialects, and can transcribe spoken words into text as well as respond to voice commands.
The API can translate text in real time or in batches, can recognize spoken words over noisy environments, and can filter out inappropriate words/content.
Google is offering the Cloud Speech API to developers for free for a limited time. Google didn’t disclose how it might price access to the API down the road.
