Introduction
Voice Mode is a powerful feature which allows you to interact with the BlackBox AI VSCode Agent using natural speech, eliminating the need for manual typing. Powered by ElevenLabs, this feature provides seamless voice-to-text and text-to-speech conversion, enabling hands-free operation for all agent capabilities.Overview
With Voice Mode enabled, you can:- Perform all tasks available through manual text input.
- Ask high-level or low-level questions about your repository
- Request code edits, new features, or debugging assistance
- Perform all chat-based interactions without using the keyboard
- Receive responses audibly, making it ideal for multitasking or accessibility
Enabling Voice Mode
To activate Voice Mode in the BlackBox AI VSCode extension:- Open the BlackBox AI chat panel in VSCode.
- Look for the voice mode button in the chat interface between the add context and the camera buttons.
- Click the voice mode button to initialize voice capabilities.
Using Voice Mode
Starting a Conversation
After enabling voice mode, you can begin speaking your queries or commands. The agent will listen and convert your speech to text.Interacting with Responses
The agent will respond both in text and audio. You can continue the conversation by speaking follow-up questions, clarifications, instructions, entirely new questions, or tasks.
Muting and Unmuting
If you need to pause voice input, use the mute button.
Capabilities
Voice Mode supports all standard BlackBox AI VSCode Agent features:- Code Analysis: Ask questions about code structure, dependencies, or functionality
- Code Editing: Request modifications, refactoring, or new implementations
- Debugging: Describe issues and request debugging assistance
- Feature Development: Propose and implement new features
- Repository Insights: Inquire about project history, commits, or documentation
- General Assistance: Any task you can perform via text chat
Examples
Requesting Code Edits
Speak: “Add capability to handle multiple user requests at once in this Java file” The agent will analyze the code and propose changes, which you can confirm verbally.
Summarizing the Code Repository
Speak: “What databases are being used in this project?” The agent will analyze the code repository and explain the contents both through voice and via text on screen.
Debugging Assistance
Speak: “Why is the API returning 500 errors? Check the logs.” The agent will investigate and provide explanations and fixes.High-Level Planning
Speak: “Design a new user authentication system for our app” The agent will outline the architecture and implementation steps.Best Practices
- Speak clearly and at a moderate pace for optimal recognition
- Use specific terminology when referring to code elements or files
- Confirm complex requests by repeating key details
- Utilize the mute function when not actively speaking to avoid accidental inputs
- Review text transcripts for accuracy before proceeding with actions
Ending Voice Mode
To disable voice mode or to end the current task, click the hang-up/end session button.
Summary
Voice Mode is a powerful feature which transforms your interaction with BlackBox AI VSCode Agent, providing a fully voice-driven development experience. Whether you’re brainstorming ideas, debugging code, or implementing features, voice mode ensures you can work efficiently without keyboard constraints. For more information on BlackBox AI VSCode Agent capabilities, refer to the Introduction and Key Features pages.Try it yourself by installing the BlackBox AI VSCode extension. Follow the installation steps to get started quickly.