Home / Agents · Audio / Hey Spark

Hey Spark
For a laptop voice assistant: local wake word and speech recognition, any LLM, MCP tools.
What is Hey Spark
Hey Spark is an open-source voice assistant for laptops by ContextuAI, with a local wake word and local speech recognition. It can use Claude or any OpenAI-compatible model as the brain and connect to tools through MCP servers. Write actions require spoken confirmation; the repository's license is marked NOASSERTION.
How to use
- Make sure you have macOS 14+ on Apple Silicon or Windows 11, a microphone, about 1 GB of disk, and a brain: Claude Code signed in (or an Anthropic API key), an OpenAI-compatible key, or Ollama running locally.
- Download the latest build from GitHub Releases: the .dmg for macOS (drag HeySpark into Applications) or HeySpark-Setup-.exe for Windows.
- Open the app and allow microphone access.
- Click the brain button, pick a provider and enter your key (or pick Ollama).
- Press Start;
- the first start downloads the speech models (about 200 MB).
- Then say "Hey Spark" and ask a question or give a command.
- Optionally add MCP connectors (GitHub, Gmail, Calendar, your own systems), shortcuts and a personality in the app's windows or config.yaml.
Use cases
- Asking a hands-free voice assistant questions on your laptop Checking your calendar or email by voice through MCP connectors Setting timers and running shortcut phrases such as a morning briefing Summarizing GitHub pull requests or starting a new project by voice, with spoken confirmation before changes Running a private voice assistant with local wake word, speech recognition and a local Ollama model
Price from official sources
Free
Check the official site for current pricing · Price not yet checked by our editors
Source: github.com · Info as of Nov 2, 2026
Company
- Name
- ContextuAI
Reviews
No reviews yet.
Alternatives
-
sweet-search
Free
For agents searching codebases: local semantic search that saves 5–33% per agent run
- Added
- Nov 2, 2026
Same category: Agents
-
LFM3-Omni-2B (Liquid AI)
Free
For on-device multimodal apps: a 2B omni-modal model supporting speech, images, and video
- Added
- Nov 2, 2026
Same category: Agents
-
nonce Not stated For securely granting AI access to your email: issues time-limited, least-privilege access per question
- Added
- Nov 2, 2026
Same category: Agents
-
Skep Free For collaborating with coding agents across devices: share live sessions with a join code
- Added
- Nov 2, 2026
Same category: Agents
Social