🩺 Vitals
- 🟢 Last active: 2026-07-18
- 📦 Latest release: 1.10.1 (2026-07-07)
- 🐞 Open issues: 13
- 🌟 Stars: 943
What do these metrics mean?
- Last active: when code was last pushed, as of our last check. The dot is green when that was recent, grey otherwise. A long gap can mean a tool is finished and stable, not only unmaintained.
- Latest release: the most recent tagged, packaged version the maintainers published. Not every healthy project tags releases.
- Open issues: unresolved reports and requests. A high number is normal for a popular project and is not a warning on its own.
- Stars: how many people bookmarked the project on its forge. A rough popularity signal, not a measure of quality.
🏗️ Profile
- Official: murmure.app
- Source: github.com/Kieirra/murmure
- License: AGPL-3.0
- Deployment: Desktop
- Data Model: Local only (no datastore; types into the active app)
- Jurisdiction: France 🇫🇷, EU 🇪🇺 (Individual Maintainer)
- Compliance (SaaS): N/A
- Compliance (Self-Hosted): Self-Hosted (User Managed)
- Complexity: Very Low (1/5) - Desktop installer, no setup
- Maintenance: Low (1/5) - Self-contained desktop app
- Enterprise Ready: Low (2/5) - No central management; compliance inherited from the endpoint
1. The Executive Summary
What is it? Murmure is a privacy-first desktop dictation application that turns speech into text entirely on-device. Press a hotkey, speak, and the transcribed text is typed straight into whatever window is focused — no audio and no text ever leaves the machine. It runs NVIDIA's Parakeet speech models locally and can optionally route transcripts through a local Ollama LLM for cleanup such as punctuation and formatting. There is no cloud, no telemetry, and no account; it is developed in France under AGPLv3 and funded entirely by community sponsorship.
The Strategic Verdict:
- 🔴 For orgs needing centrally managed, supported dictation at scale: Caution. Murmure is a single-maintainer project with no central management, SSO, or vendor support contract. Treat it as an endpoint utility, not a managed service.
- 🟢 For privacy-sensitive or regulated desktop users (legal, medical, public sector): Strong Buy. Because transcription is 100% local with zero telemetry, the workflow inherits the compliance posture of a hardened endpoint — sensitive dictation never traverses a third-party cloud like a US-hosted dictation SaaS.
2. The "Hidden" Costs (TCO Analysis)
| Cost Component | Wispr Flow (SaaS) | Murmure (Self-Hosted) |
|---|---|---|
| Subscription | Per-user monthly subscription for cloud dictation | Free and open source; no subscription or per-seat cost |
| Audio Processing | Audio streamed to and processed in the vendor cloud | 100% on-device; audio and transcripts never leave the machine |
| Model & Compute | Managed in the vendor's cloud | Local Parakeet models on your own CPU/GPU; optional local Ollama for cleanup |
3. The "Day 2" Reality Check
🚀 Deployment & Operations
- Installation: Native desktop installers — Windows MSI, Linux
.deb, and macOS.dmg. No server, account, or internet connection is required; the speech models run locally, with optional Ollama integration for LLM post-processing. - Scalability: Per-device by nature. "Rollout" means packaging the installer and the chosen Parakeet and Ollama models through your existing endpoint or MDM tooling; there is no central server to operate.
🛡️ Security & Governance (Risk Assessment)
- Jurisdiction & Geopolitics: Maintained in France under EU jurisdiction (GDPR), but because Murmure collects no data and runs fully offline, data-residency and surveillance concerns — including the US CLOUD Act — are inapplicable. There is no vendor backend to subpoena.
- The Compliance Shift: With no cloud processing, the compliance burden moves entirely to the endpoint. If the device is compliant (encrypted disk, HIPAA-aligned MDM), the dictation workflow inherits that posture — Murmure introduces no new third-party processor into your data map.
- License & Governance Risk: AGPLv3 is benign for desktop dictation, but its network-copyleft clause bites if you wrap Murmure's experimental API into a networked transcription service, which could force you to open-source the wrapper. The larger caution is governance: a single maintainer funded by donations carries real abandonment risk, mitigated mainly by the trivial exit cost — no datastore, just a standard desktop app.
4. Market Landscape
🏢 Proprietary Incumbents
- Wispr Flow: Modern AI dictation sold as a per-seat subscription. Fast and polished, but audio is streamed to and processed in its cloud.
- Dragon (Nuance): The long-standing enterprise dictation incumbent for legal and medical work, but expensive, Windows-centric, and now Microsoft-owned.
🤝 Open Source Ecosystem
- OpenWhispr: The closest peer — local hotkey dictation that also runs Whisper and Parakeet on-device. OpenWhispr is MIT with paid cloud, agent, and SSO tiers; Murmure is fully free under AGPL, EU-maintained, with optional local Ollama cleanup.