Ask questions about your data without typing a single character. Voice input lets you dictate messages, review the transcript, and send — useful for quick queries while multitasking or when a full keyboard is inconvenient. It is available on Web, Desktop (macOS), and Mobile; the Windows desktop app does not include voice input.
Voice input uses your device's built-in speech recognition to transcribe what you say. The transcribed text is inserted into the chat input for review before sending — nothing is sent automatically.
Dictate a message hands-free and confirm the transcript before sending, reducing friction for quick questions like "How many rows in the pageviews table?" or "Show me the top audiences by size."
- Treasure AI Studio on Web, Desktop (macOS), or Mobile
- Microphone permission granted to Treasure AI Studio in your browser or device settings
Select the microphone button in the chat input toolbar (next to the Send button) to open the voice input interface, speak your message, review the real-time transcript, then confirm to insert the text into the chat input. Confirming inserts the transcript without sending, so you can edit before you send. The interaction is the same on Web, Desktop (macOS), and Mobile.
- Tap the microphone button in the chat input toolbar (next to the Send button)
- The voice input interface opens, replacing the standard text input
- Speak your message — Studio transcribes your speech in real time
- Review the transcript displayed at the bottom of the voice input area
- Tap the checkmark button to confirm and insert the text into the chat input
- Tap the X button to cancel and return to the standard input without inserting any text
When voice input is active, you see:
| Element | Description |
|---|---|
| Waveform visualization | A 50-bar audio waveform that responds to your voice level in real time |
| Recording indicator | A pulsing red dot that indicates active recording |
| Elapsed timer | Shows how long you have been recording (e.g., "0:05", "1:23") |
| Transcript preview | Your transcribed speech appears at the bottom, updated as you speak (up to 2 lines visible) |
| Confirm button | Checkmark icon — inserts the transcript into the chat input |
| Cancel button | X icon — discards the recording and returns to the standard input |
When you tap the checkmark button:
- The transcribed text is inserted into the chat input area
- If you already had text in the input, the voice transcript is appended to the existing text (separated by a space)
- The voice input interface closes and the standard chat input returns
- You can edit the inserted text before sending, or send immediately
Voice input does not automatically send the message. You always have a chance to review and edit before tapping Send.
Voice input requires microphone access. The first time you use it, your browser (Web) or device (Desktop, Mobile) prompts you to grant microphone permission to Treasure AI Studio.
If microphone permission was previously denied, re-enable it for Treasure AI Studio:
- Web — Open the site's permission settings in your browser (in most browsers, from the icon in the address bar), allow the microphone, then reload the page.
- Desktop (macOS) — Allow microphone access for Treasure AI Studio in your operating system's microphone privacy settings, then restart the app.
- iOS — Open Settings > Treasure AI Studio, enable Microphone, then return to the app and try voice input again.
Voice input transcribes in the language set for the app — English or Japanese. Set it in the app's language setting before dictating; the recognizer uses that language, not a separate device-level locale.
The app records audio with real-time metering (which drives the waveform visualization) and passes it to the device's speech recognition to produce the transcript.
- Tap the microphone and see the waveform respond to your voice
- See a real-time transcript appear at the bottom of the voice input area
- Tap the checkmark and see the transcript inserted into the chat input field
| Issue | Solution |
|---|---|
| Waveform doesn't respond to your voice | Check that microphone permission is granted for Treasure AI Studio — on iOS, in Settings > Treasure AI Studio > Microphone; on Web and Desktop, in your browser or operating system settings. If permission was recently changed, reload the page (Web) or restart the app (Desktop, Mobile) |
| Transcript is empty or inaccurate | Speak clearly at a natural pace. Background noise reduces accuracy. Speech recognition works best in quiet environments |
| Voice input button not visible | Voice input is not available on the Windows desktop app, so no microphone button appears there. On other platforms, make sure you are on a recent version of the app (Desktop, Mobile) or have reloaded the page (Web). If the microphone button still doesn't appear, check that microphone permission is granted in your browser or device settings |
Combine voice input with typing. Start typing a complex message, then tap the microphone to dictate the rest — the transcript appends to your existing text.
- Getting Started: Set up Studio on Web, Desktop, or Mobile
- AI Chat Interface: Full chat interface guide
- Chat Persistence & History: How conversations are saved across devices