Voquill provides a cross‑platform AI voice dictation engine that runs locally on macOS, Windows, and Linux, allowing users to dictate into any text field without internet dependence. The system uses open‑source transformer models with optional API keys for LLM inference and includes neural post‑processing to remove filler words and autocorrect text. Subscription tiers add encrypted cross‑device sync, unlimited transcription, and enterprise on‑premise deployment with API integration.
Funding
Funding not disclosed
Founders
Product
Problem
Typing speed limits productivity for knowledge workers, and existing dictation tools often route audio to proprietary cloud services, exposing sensitive content and requiring constant internet connectivity. This creates a barrier for users who need fast, private, and reliable voice-to-text across diverse desktop environments.
Solution
Voquill delivers a cross‑platform AI‑driven voice dictation engine that operates system‑wide on macOS, Windows, and Linux, allowing users to dictate into any text field without switching applications. The core speech‑to‑text pipeline runs locally on the user’s device, leveraging open‑source transformer models and optional user‑provided API keys for LLM inference, ensuring that audio never leaves the machine unless the optional cloud service is enabled. An integrated post‑processing module applies neural autocorrect to strip filler words, false starts, and hesitations, producing clean, publish‑ready text. For power users, Voquill offers cross‑device synchronization, unlimited monthly transcription quotas, and priority support via a subscription tier, while enterprise customers can deploy the solution on‑premises and integrate custom workflows through a documented API. The open‑source architecture and transparent privacy model give organizations full control over data residency and compliance.
Target Audience
The primary users are individual professionals and remote workers who need fast, private dictation across multiple desktop OSes, while enterprises and regulated organizations adopt the on‑premise version for secure, compliant voice‑enabled workflows.
Features
- System‑wide input hook that captures voice in any application window on macOS, Windows, or Linux
- Local inference engine using open‑source transformer models; optional “bring‑your‑own” API key support for external LLM providers
- Offline‑first operation with GPU/CPU acceleration; no mandatory internet connection for core transcription
- Neural post‑processing layer that removes filler words, hesitations, and performs context‑aware autocorrect
- Cross‑device sync and unlimited word quota in the Pro subscription, stored encrypted on the user’s cloud of choice
- Enterprise‑grade on‑premise deployment package with customizable integration hooks and compliance‑focused data handling
- Fully open‑source codebase with community‑driven extensions and transparent security audits
- API and SDK for embedding dictation capabilities into proprietary software stacks