whereismyai
Hermes Whatsapp Voice
Hermes WhatsApp Voice handles live WhatsApp voice calls for Hermes bots using Gemini Live and a Rust sidecar.
Hermes WhatsApp Voice enables live voice calls between users and Hermes bots over WhatsApp. A Rust sidecar links to a separate WhatsApp number as a device and bridges call audio to Gemini Live for speech-to-speech communication. The voice model uses tools to execute Hermes tasks mid-call and ends calls on command. When a call concludes, Hermes generates and sends a text summary in WhatsApp. The system runs on Linux and macOS, requiring a Hermes instance, a bot WhatsApp number, and a Gemini API key.
Hermes Whatsapp Voice at a glance
- Best for: Software engineer
- Runs on: Linux, macOS, self-hosted, open source
- Pricing: Free plan; paid use roughly $0.005 per minute of audio in and $0.018 per minute out for Gemini Live
- Released: October 2026
- Category: AI agent tool
- Website: github.com/ashwanthbalakrishnan5/hermes-whatsapp-voice
Key features
- Call the bot on WhatsApp for live speech interaction
- Let Hermes initiate calls for urgent reminders or alerts
- Generate and send text summaries after calls end
- Execute Hermes agent tools mid-call via the askhermes tool
- Restrict calls strictly to the designated owner number
- Bridge call audio to Gemini Live for speech-to-speech
Use it for
- Talk to a Hermes bot over live WhatsApp voice calls
- Receive phone calls from Hermes for urgent reminders
- Manage calendar events and tasks using voice commands
- Get post-call text summaries of conversations in WhatsApp
Alternatives to Hermes Whatsapp Voice
- Hermes WebUI — Run one server and access your agent from any device.
- Hermes Agent — It features persistent memory, self-created skills, and isolated sandboxing.
- Hermes Desktop — It features a built-in learning loop that creates and improves skills from experience.
- Hermes Gadget — Brings a personal Hermes Agent to a physical desk with voice interaction.
- Hermes Browser Extension — Connects active web context directly to the local Hermes runtime.
- Hermex — Native mobile access to a self-hosted Hermes agent with no subscription fees.
- Whatsapp MCP — Read-only personal WhatsApp access hosted entirely in your Cloudflare account.
- Herald OS — It runs your entire machine through conversation with an AI agent.
- Buddylist — Manage real coding agents and group chats in a classic messenger style.
- Oh My Hermes — It adds an operating layer to Hermes Agent with explicit evidence boundaries.
- Memory OS for Hermes — Stops agents from forgetting by adding local persistent memory layers.
- Hermes Atlas — Curates over 250 open-source tools and skills with live GitHub data.
All Hermes Whatsapp Voice alternatives