Voquill

Voquill is an open-source voice reporting tool specifically designed for pathologists, offering a free alternative to proprietary dictation solutions like WisprFlow. It is a medical-grade dictation solution that converts natural speech into structured, formatted pathology reports in real time, allowing pathologists to type four times faster than manual typing. Backed by Y Combinator, Voquill uses agentic speech recognition that understands the context of pathology reporting. The core value is that pathologists can keep their eyes on the slide and speak naturally while Voquill listens in the background, drafts the report, and populates each field automatically. This eliminates the need to switch between applications or manually edit dictation results, allowing pathologists to sign out reports faster.

Traditional pathology reporting often involves repetitive, time-consuming steps: dictating findings, then manually clicking through fields to place each observation, and finally editing for formatting and phrasing. Pathologists may spend as much time editing as they do examining slides. Voquill solves this by using agentic speech that recognizes not just words but the structure of a synoptic report. It automatically routes each dictated finding to the correct field—specimen, microscopic, interpretation, comments—based on the natural flow of speech. This directly addresses the pain point of clicking around to find the right field and reduces cognitive load, allowing pathologists to focus on diagnosis rather than data entry. Voquill works system-wide across macOS, Windows, and Linux, fitting into any pathologist's preferred environment.

A standout feature of Voquill is its agentic field placement capability, also described as "field-aware report fill". When a pathologist speaks, Voquill identifies which part of the report each statement belongs to—for example, differentiating between a specimen description and a microscopic finding. It then inserts that text into the appropriate field within the synoptic report template. This works because Voquill has been trained on pathology language and report structures. The benefit is that pathologists do not need to manually select fields or use mouse clicks; they simply talk through the case, and every word lands in the right place. This feature is explicitly mentioned on the site as "Every word in the right field, automatically," and it saves significant time during dictation by eliminating field navigation and reducing data entry errors.

Voquill distinguishes itself by learning each pathologist's personal writing style. It reads the user's old reports to adopt their preferred phrasing, conventions, and terminology, ensuring that new reports sound consistent with prior work. This reduces the need for post-dictation editing. Additionally, Voquill supports customizable snippets—reusable blocks for phrases used frequently, such as "GI biopsy", "breast excision", or "prostate core". These snap into place when the pathologist uses the corresponding voice trigger. The combination of style learning and snippet functionality means that pathologists can dictate in their own natural language, and Voquill will produce reports that read exactly as they would have written them. Users can also bring their existing templates and snippets from their lab, as Voquill plugs straight into them.