GitHub invites programmers to speak to the co-pilot directly

Briefly GitHub is testing a new feature that will allow developers to instruct the AI-powered programming assistant Copilot to generate code using voice commands.

The legally disruptive and annoying technology of licensing is not a simple speech-to-text dictation engine that requires developers to read their software source line by line. Alternatively, you can use the “Hey, GitHub!” button. It acts as a voice interface for Copilot, which automatically suggests a code from the prompts.

Hopefully, programmers can say things out loud like generic job descriptions, and ask Microsoft-owned Copilot to recommend the source to meet that demand.

As usual, developers can decide if they want to keep or cancel KoPilot’s suggestions. Hi Github! It is designed to help them program faster using their voice. They can instruct the program to autocomplete the standard code, and manually edit any suggested output using their keyboards. The new feature can also be used to convey code or provide summaries to make scripts easier to read and understand.

Hi Github! It will be provided as part of the $10 subscription fee for Copilot. If you are interested you can sign up for a technical preview here

The new Amazon AI bot

Amazon has demonstrated a new robotic arm, called Sparrow, that runs machine learning algorithms to automatically identify and sort items for packaging. Useful where Amazon workers try to join unions or complain about working conditions and long working hours.

Sparrow was introduced on stage at Amazon’s Future Delivery Conference this week. It is a large L-shaped lever with a clutch at one end; Suction cups at the end of the gripper fingers are used to pick up items and sort them into boxes. Sparrow can successfully capture all kinds of objects of different sizes, even if they have curved surfaces, said Jason Messinger, chief technical product manager for robotic manipulation at Amazon Robotics.

Using computer vision technology, the computer system that controls the robotic arm is able to recognize objects and can identify about 65 percent of Amazon’s inventory. “This is not just about picking and moving the same objects with high precision, which we’ve seen in previous robots,” Messinger said, according to CNBC.

Amazon is investing in AI bots to perform tedious and repetitive tasks, potentially relieving itself of the need to hire too many humans.

Midjourney releases AI text-to-image tool

Midjourney, best known for creating subscription-based text-to-image software, especially artistic software, has released its fourth edition.

“V4 is a completely new database and a completely new AI architecture,” Midjourney founder David Holz said on the company’s Discord channel. “It’s our first model that has been trained on a new Midjourney AI suite and has been working on it for over 9 months.”

The folks at Ars Technica tested the model and saw an improvement in v4’s ability to convert text prompts into images that feature better scene compositions and produce more appropriate sizes of objects relative to each other, compared to v3. The latest version was also better at producing more realistic images.

Halls said previously record He didn’t want Midjourney to be too good at creating photos that are realistic enough to pass them off as fakes. “For us, when we were refining it, we wanted it to look kind of pretty, and beauty doesn’t necessarily mean realism.

“If anything, we’re actually skewing it a little bit further away from images…I know this technology can be used as a deep superhero machine. And I don’t think the world needs more fake photos. I don’t really want you to be the source of fake photos in the world.” . ®

#GitHub #invites #programmers #speak #copilot

Leave a Comment

Your email address will not be published. Required fields are marked *