Posts

Showing posts from October, 2025

[MS] Microsoft C++ Team at CppCon 2025: Trip Report - devamazonaws.blogspot.com

Image
Another year and another CppCon is all wrapped up! Microsoft was pleased to once again be an exhibitor and to have several of our employees present sessions at the conference. Here are my major takeaways from the keynotes, sessions, conversations, and the general vibe. AI tooling is finding a footing with the C++ community. The next year is going to see the reliability and usefulness of those tools jump drastically. Visual Studio users consistently name the debugger as one of the best features of the product. Lots of folks are getting excited about the static reflection capabilities in C++26. The week before CppCon, we announced Visual Studio 2026 and Microsoft C++ (MSVC) Build Tools v14.50, and the conference gave us a great opportunity to let you know what you can expect in the new release and for you to give us your feedback. We highly encourage you to try out the Insiders release and let us know what your experience is like. This is a long post, so here’s a table of contents ...

[MS] Tuning and Optimization of Speech-to-Text (STT), Text-to-Speech (TTS), and Custom Keyword Recognition in Azure Speech Services - devamazonaws.blogspot.com

Introduction Optimizing Speech-to-Text (STT) and Text-to-Speech (TTS) is essential for developers building voice-enabled applications, as it enhances recognition accuracy and overall user experience. Additionally, custom keyword recognition enables hands-free activation for voice assistants. This blog outlines: Methods for improving STT accuracy and tuning How to enhance TTS pronunciation using custom lexicons Approaches for implementing custom keyword recognition efficiently Speech-to-Text (STT) Optimization Obtaining STT Models Azure provides several STT models: Default STT Models : Available via Azure AI Foundry portal for general-purpose speech recognition. Custom Speech Models : Trainable via Azure AI Foundry portal for specific accents and terminology. Embedded Speech Models : Designed for edge deployment, enabling low-latency offline processing . This is a Limited Access offering and applications can be submitted via Azure AI Speech embedded speech limited access re...

Amazon WorkSpaces announces USB redirection support for DCV WorkSpaces - devamazonaws.blogspot.com

AWS announces USB redirection support for WorkSpaces running Amazon DCV protocol, enabling users to access locally connected USB devices from their virtual desktop environments. With this feature, customers can now connect a wide range of USB peripherals to their virtual desktops, including credit card readers, 3D mice, and other specialized devices. USB redirection addresses the need for direct access to USB devices that require specialized drivers or lack dedicated protocols. This capability is currently limited to WorkSpaces Personal with Windows desktops accessed from Windows client devices. Performance and device compatibility may vary, so testing with your specific USB peripherals is recommended before adding them to the allowlist. This feature is available in all AWS Regions where Amazon WorkSpaces is offered. For more information about USB redirection in Amazon WorkSpaces, see USB Redirection for DCV in the Amazon WorkSpaces Administration Guide , or visit the Amazon Work...

AWS Elastic Beanstalk adds support for Amazon Corretto 25 - devamazonaws.blogspot.com

AWS Elastic Beanstalk now enables customers to build and deploy Java applications using Amazon Corretto 25 on Amazon Linux 2023 (AL2023) platform. This latest platform support allows developers to leverage the newest Java 25 features while benefiting from AL2023's enhanced security and performance capabilities. AWS Elastic Beanstalk is a service that provides the ability to deploy and manage applications in AWS without worrying about the infrastructure that runs those applications. Corretto 25 on AL2023 allows developers to take advantage of the latest Java language features including compact object headers, ahead-of-time (AOT) caching, and structured concurrency. Developers can create Elastic Beanstalk environments running Corretto 25 through the Elastic Beanstalk Console, CLI, or API. This platform is generally available in commercial regions where Elastic Beanstalk is available including the AWS GovCloud (US) Regions. For a complete list of regions and service offerings, see...

[MS] Visual Studio October Update – new models, memories, planning, and more - devamazonaws.blogspot.com

Image
The October 2025 update for Visual Studio 2022 (v17.14) is now available. In this month, we are bringing you improvements to model choices, agentic flows, and chat thread management.     New Models   We now have  Claude Sonnet 4.5  and Claude Haiku 4.5 available in the chat window. This means the latest innovations for driving your agentic workflows is right at your fingertips.   Memories   For Copilot to work effectively for your unique team and repository, it needs to learn about your project and team best practices. Copilot memories enable Copilot to understand and apply your project's specific coding standards, making it project-aware and consistent across sessions.   How Copilot memories work  Memories use intelligent detection to understand your team's preferences as you prompt in the chat. That means, as you prompt how you normally would, Copilot is looking for insta...

[MS] Dev Proxy v1.3 with exporting to HAR, LLM usage tracking, and enhanced permissions analysis - devamazonaws.blogspot.com

Image
We're thrilled to announce the release of Dev Proxy v1.3.0! This update introduces two powerful new plugins: HAR file generation for universal debugging and LLM usage tracking for cost optimization, along with enhanced permissions analysis and improved OpenAPI spec generation for complex testing scenarios. In this version: New HarGeneratorPlugin for HTTP Archive file generation New OpenAIUsageDebuggingPlugin for language model usage tracking Enhanced minimal permissions plugins with authentication scheme support Improved OpenAPI spec generation for multi-port scenarios Better CSV logging for API activity tracking Bug fixes and improvements New HarGeneratorPlugin for HTTP Archive file generation We're introducing the HarGeneratorPlugin , a powerful new addition that generates industry-standard HTTP Archive (HAR) files from intercepted API requests and responses. What are HAR files? HTTP Archive files are the universal standard for recording network activity. They'...

[MS] What to do when you have a crash in the runtime control flow guard check - devamazonaws.blogspot.com

Windows Control Flow Guard (CFG) is a defense in depth feature which validates indirect call targets. The idea is that each module that is enabled for CFG provides a bitmap that describes which addresses in the module are intended to be targets of indirect calls. When CFG is enabled in a process, indirect function calls are checked against this table, and if the address is deemed invalid, the process terminates itself, and the Watson service records the details for future investigation. If you are studying a crash in the control flow guard validator¹ you may want to pick out the failed address so you can understand better what went wrong and use it to guide the next step of your debugging. (Was it a bad address? Was the DLL unloaded? Was it a garbage value due to use-after-free?) In general, the control flow guard validator takes a function address in some register, performs shifting and masking operations using that register as a source (to calculate the bit position in the call tar...

[MS] New GitHub Copilot capabilities for C++ developers: Upgrade MSVC, improve build performance, and refactor C++ code - devamazonaws.blogspot.com

Image
[alert type="important"]GitHub Copilot app modernization for C++, GitHub Copilot build performance for Windows, and C++ code editing tools for GitHub Copilot will be available soon in Private Preview. Join the waitlist for Private Preview today .[/alert] Yesterday at GitHub Universe, GitHub announced support for custom agents. Custom agents are specialized configurations with custom prompts and context that focus Copilot on specific development tasks. On the C++ team, we have a similar goal: Provide differentiated capabilities that optimize GitHub Copilot for C++ development tasks. As a first step towards this goal, we’re announcing new capabilities for GitHub Copilot that support C++ developers in three development tasks: Upgrade Microsoft C++ (MSVC) Build Tools Improve C++ build performance Perform large-scale C++ refactors These experiences will be available soon as a Private Preview in Visual Studio 2026. Join the waitlist for Private Preview today . Upgrade MSVC w...

[MS] Deprecation of the previewImageUrl property for Insights items that reference file attachments - devamazonaws.blogspot.com

We’re deprecating the previewImageUrl property in the   resourceVisualization complex type for Insights items  where the referenced resource is a file attachment ( resourceReference.type  = fileAttachment ). This affects responses from the Insights List shared  and Insights List used  APIs when the underlying resource is an email attachment. What’s changing and why The previewImageUrl property is no longer returned for file attachment items. File attachments don’t have a consistent preview experience across Microsoft 365, so removing this property simplifies the API and aligns it with supported preview scenarios. Items that reference cloud files (like driveItem ) aren't impacted. Timeline Deprecation notice : Effective immediately Removal date : November 1, 2027 Action required If your app uses resourceVisualization.previewImageUrl for file attachment insights, update your logic to handle the absence of this property. For previews, prefer modern a...

Web Grounding: Build accurate AI applications with Amazon Nova models - devamazonaws.blogspot.com

We are excited to announce the general availability of Web Grounding, a new built-in tool for Nova models. Customers can use Web Grounding today with Nova Premier using the Amazon Bedrock tool use API. Support for additional Nova models is coming soon. Web Grounding is a built-in tool that can be used to retrieve and incorporate publicly available information with citations as context for responses. Developers can use the Web Grounding tool to implement a turnkey Retrieval Augmented Generation (RAG) solution using current, real-time information, reducing hallucinations and leading to more accurate outputs. Web Grounding is available today in US East (N. Virginia), US East (Ohio), and US West (Oregon) AWS Regions via cross-region inference. Learn more about using the Web Grounding tool on Nova models and steps to get started at our blog post . Post Updated on October 29, 2025 at 01:00PM

[MS] MCP Registry and Allowlist Controls for Copilot in JetBrains and Eclipse Now in Public Preview - devamazonaws.blogspot.com

Image
MCP registry and allowlist controls for GitHub Copilot in JetBrains IDEs and Eclipse are now available in public preview in nightly/pre-release builds. What’s new MCP Registry An MCP Registry is a directory of Model Context Protocol (MCP) servers. For users of JetBrains IDEs and Eclipse, you can now configure your MCP Registry and browse available MCP servers directly within your IDE. This greatly streamlines setup and provides a seamless experience for discovering and managing MCP servers right from the editor. Allow List Controls As an enterprise or organization owner, you can configure an MCP Registry URL along with an access control policy. These settings determine which MCP servers your developers can see and run in supported IDEs with GitHub Copilot. When combined with the Registry only policy , it prevents any usage of MCP servers (at runtime) that are not defined in the internal registry. Set up your MCP Registry In JetBrains IDEs:  Sign in and open Copilot chat, then...