ML Kit expands into NLP with Language Identification and Smart Reply
Blog post from Firebase
ML Kit has introduced two on-device natural language processing APIs, Language Identification and Smart Reply, expanding the platform beyond its prior image and video-focused capabilities for iOS 9.0+ and Android 4.1+. Smart Reply generates suggested responses from up to 10 recent chat messages through a stateless process that neither retains conversation history nor sends it to servers, initially supporting English and incorporating safeguards to avoid suggestions for sensitive topics or unsupported languages. Built with TensorFlow Lite, SentencePiece encoding, and Transformer-based models, it is designed for simple integration into messaging applications and has been adopted by textPlus. Language Identification recognizes text across 110 languages, usually using only a few words and returning results in roughly 1–2 milliseconds, with BCP-47 language codes, an undetermined result when confidence is insufficient, and optional confidence-ranked language alternatives.
No tracked trend matches for this post yet.
Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.