“The best part about machine learning is deep learning, the more we understand you the more accurate the platform gets.” – Kavin Bharti Mittal
As of now, theMessenger Whats Stickers app has over 30K stickers available in over 40 Indian languages and dialects, which Mittal says only covers about 15% of vocabulary for online communications. The company’s idea is to use AI and ML to fill the wide gap in demand and supply of multilingual stickers.
Mittal underlines this as the USP of its revamped app.whatsapp “We are able to give you the best stickers because we understand your age, your gender, your location, who you are, your personal preferences, the season of the year, and we broadly understand Messenger Whats Stickers the relationship between our users based on the chats,” Mittal said.
Mittal said itMessenger Whats Stickers would help the AI tailor sticker recommendations. “A lot of the content we will create now which will take us from 30K stickers to around 100K will be mostly machine generated and that is really powerful because that lets us scale up languages,Messenger Whats Stickers” he added.Friend 1: Hi. Where are you? Friend 2: On my wayFriend 1: How long will you take to reach?Friend 2: Five minutes away.Friend 2: Buy the tickets!
It’s a real conversation; which one can have on WhatsApp or Facebook Messenger or plain Messenger Whats Stickersold SMS. But the interesting part is that Mittal didn’t type out an actual word, nor did he use voice messages for the chat; instead, he used stickers. The whole conversation, with those three to four phrases per side was carried out on stickers.
9Apps 4.9
