Voice Input & Audio Messages

Sidekick v1.0.60 adds voice input to the chat interface, letting you speak to your agent instead of typing. The agent transcribes your recording and responds in the same conversation thread.

Using Voice Input

Press-and-Hold to Record

1. Open the Sidekick chat interface (dashboard or mobile browser).
2. Press and hold the microphone icon next to the message input field.
3. Speak your message clearly.
4. Release the button to stop recording.
5. The agent automatically transcribes your audio and generates a response.

Tip: Voice input is particularly well-suited to mobile devices where typing long instructions is cumbersome.

Forwarding Voice Notes from Telegram or WhatsApp

If you receive or record a voice note in Telegram or WhatsApp, you can forward it directly to your connected Sidekick agent:

1. In Telegram or WhatsApp, long-press the voice note you want to forward.
2. Select Forward and choose your Sidekick agent as the recipient (via the connected integration).
3. The agent receives the audio, transcribes it, and processes it as a normal instruction or query.

This means spoken reminders, requests, or notes captured in messaging apps can flow straight into your agent workflow without any manual transcription on your part.

Supported Platforms

Platform	Press-and-Hold Recording	Forwarded Voice Notes
Sidekick Web (desktop)	✅	—
Sidekick Web (mobile)	✅	—
Telegram integration	—	✅
WhatsApp integration	—	✅

Notes & Limitations

• Transcription is handled automatically on the Sidekick cloud — no local software or third-party transcription account is required.
• Transcription accuracy depends on audio clarity and background noise. Speak clearly and close to your device microphone for best results.
• Very long recordings may take a moment to process before the agent responds.
• The original audio is not stored after transcription; only the transcribed text is retained in your conversation history.