Voice Mode

Talk to OpenCharts using real-time streaming voice. Voice Mode uses Gemini Live for low-latency, natural conversation — and everything you can do in text chat works with voice too.

Getting Started

Open Chat from the sidebar, then either:

  • Click the microphone button next to the send button, or
  • Switch to Voice mode using the mode pills (keyboard shortcut ⌘5), then click the mic.

Your browser will ask for microphone permission the first time. Once granted, the voice overlay opens and you'll see the animated orb — start speaking naturally.

The Voice Orb

The animated orb at the center of the voice overlay shows you what's happening:

Idle

Gentle breathing animation — waiting for you to speak.

Listening

Audio-reactive rings respond to your voice level.

Thinking

Rotating glow — the AI is processing your request.

Speaking

Pulsing rings — the AI is talking back to you.

What You Can Do

Voice Mode works with all the same tools as text chat. Speak naturally and the AI will figure out what to do.

Create Projects

Say "create a flowchart for user onboarding" and it builds one instantly.

Deep Research

Ask to "research the latest React patterns" — triggers web search and synthesis.

Quick Answers

General questions get fast, conversational responses with no extra steps.

Image Generation

Describe an image and it generates one, just like in text mode.

Controls

Mute / Unmute

Temporarily mute your microphone without ending the session.

End Session

Click the red X button or say goodbye to close voice mode.

Voice Settings

Click the gear icon next to the mic button to pick a voice and microphone.

Microphone Selection

If you have multiple audio input devices (USB microphones, headsets, AirPods, etc.), you can choose which one Voice Mode uses.

  • Click the gear icon () next to the microphone button in the chat input area.
  • The Microphone section at the top of the settings popover lists all detected audio input devices.
  • Your system default microphone is automatically selected. Tap any device to switch.
  • Your choice is saved and will be used for all future voice sessions.
If no devices appear, click "Grant access" to allow OpenCharts to see your microphones. You may need to update your browser's site permissions if you previously denied access.

Switching Between Text and Voice

You can toggle voice on and off in any conversation — it doesn't have to be a new chat.

  • Text → Voice: Open any existing conversation and click the mic button (or press ⌘5). The AI reads the conversation history and acknowledges it's caught up before continuing.
  • Voice → Text: End the voice session — everything you said is saved as messages in the conversation. You can continue typing normally.
  • Resuming voice: Open a previous voice conversation and click the mic again. The AI picks up where you left off with full context.
When you switch to voice from a text conversation, the AI will briefly confirm it's up to speed — something like "I'm caught up on our conversation, what's next?" You don't need to repeat anything that was already discussed.

Live Transcripts

As you speak, a real-time transcript appears below the orb showing what was said. Both your words and the AI's responses are displayed so you can follow along.

When the AI creates a project or performs an action, an action card appears with a direct link to the result.

Tips for Best Results

  • Speak clearly and at a normal pace — the AI uses voice activity detection to know when you've finished.
  • Be specific: "Create a flowchart showing the login process with error handling" works better than "make a chart."
  • You can interrupt the AI while it's speaking — just start talking and it will stop and listen.
  • Use a headset or earbuds to avoid echo from the AI's audio output.
  • Voice Mode uses AI credits similar to text chat.

Requirements

  • A modern browser (Chrome, Edge, Firefox, Safari) with microphone access.
  • Microphone permission granted to OpenCharts.
  • Available AI credits on your account.
Voice Mode is fully additive — everything you already use in text chat still works exactly the same. Voice is just another way to interact with the same AI capabilities.
Was this article helpful?

Related Articles