“In most supermarkets today, bread has lost its emotional connection with the shopper,” said Randall Wilkinson, CEO of The Wilkinson Baking Company. “In the age of home delivery, The BreadBot attracts consumers back to the store because it delivers fresh, delicious bread.”
Technical specifications of the device:
- Produces 10 loaves per hour
- Time to first completed loaf is 90 minutes
- New loaves produced every 6 minutes
- Operates up to a 24 hour duty cycle
- Yields a maximum of 235 loaves per day
- Requires about 30 minutes for daily cleaning
CES: Language tech means no more “lost in translation”
Talking to strangers in foreign countries just got a lot easier with recent advancements in translation technology. Last week, major companies and small startups alike showed the CES technology expo in Las Vegas how well their translation worked at live translation.
Most existing translation apps, like Bixby and Siri Translate, are still in their infancy with live speech translation, which brings about the need for dedicated solutions like these technologies:
Babel’s AIcorrect pocket translator
The AIcorrect Translator, developed by Beijing-based Babel Technology, attracted attention as the linguistic king of the show. As an advanced application of AI technology in consumer technology, the pocket translator deals with problems in cross-linguistic communication.
It supports real-time mutual translation in multiple situations between Chinese/English and 30 other languages, including Japanese, Korean, Thai, French, Russian and Spanish. A significant differentiator is that major languages like English being further divided into accents. The translation quality reaches as high as 96%.
It has a touch screen, where transcription and audio translation are shown at the same time. Lei Guan, CEO of Babel Technology, said: “As a Chinese pathfinder in the field of AI, we designed the device in hoping that hundreds of millions of people can have access to it and carry out cross-linguistic communication all barrier-free.”
Click here to read about the Pilot, Travis, Pocketalk, Google and Zoi translators.
How Netflix gets to know you
It can sometimes be downright creepy the way streaming services like Netflix and Spotify pinpoint your tastes precisely. Obviously, much of it is based on the genres and performers previously accessed. But there is more to it, as Netflix told us in the following guide.
Whenever you access the Netflix service, the recommendation system attempts to help you find a show or movie to enjoy as quickly and easily as possible. The service bases its suggestions on a number of factors including interactions with the service (such as how you rated viewed titles), information on titles available on the service (such as the genre, release year and cast) and how they compare to your favourites, and the viewing preferences of other members who have a similar taste in entertainment.
The Netflix recommendation system does not include demographic information – like age or gender – as part of the recommendation making process.
If you can’t find something you want to watch under the recommendations, you can always search the comprehensive catalog. When you enter a search query, the top results returned are based on the actions of other members who have entered the same or similar queries. Netflix has also introduced mobile previews on your on-the-go devices, giving you a fun,and easy way to learn about new content and find what you want to watch even faster.
“Jump starting” personalised recommendations
When you create your Netflix account, or add a new profile to your account, you are asked to choose a few titles that you like. This information is used to “jump start” your recommendations, giving Netflix a starting point to work from with a basic understanding of your preferences. Identifying these titles is optional but if you choose to forego this step, you will begin your Netflix experience with a list of diverse and popular titles to choose from. Your viewing preferences from this point will then feed into the recommendation system and enable greater personalisation as time goes on.
For those who do enter some of their favourite titles to jump start recommendations, once you start watching other titles on the service, this will “supercede” any initial preferences and the titles you’ve watched more recently will outweigh titles you watched in the past in terms of driving recommendations.
Rows, rankings and title representation
In addition to carefully selecting which titles to include in the rows on your Netflix profile homepage, the recommendation system also ranks each title within each row and ranks the rows themselves – all in order of what you’re likely to enjoy most.
The most strongly recommended rows are therefore placed at the top, with the most strongly recommended titles at the beginning of each row, on the left (unless you have selected Arabic or Hebrew as your language in which case these titles will be ordered from right to left).
So, if you’ve ever thought that maybe your Netflix profile knows you better than your closest friend – you may be right. But it’s not magic, it’s just the “Netflix Effect”, personalising the entertainment experience.