Talk2Invoice: Turn a Job Note Into a Client-Ready Draft With Open-Weight AI

Description and use

Talk2Invoice enables independent workers to quickly turn a brief job note into a client‑ready invoice draft using a local LLM, keeping data on the user’s device.

Problem

Independent service workers need a quick way to convert a brief job note into a client‑ready invoice draft without learning complex accounting software.

Outcome

The prototype demonstrates that a local LLM can suggest invoice fields from a job note, reducing manual entry time and keeping data on the user’s device.

Workflow

• 1. User types or dictates a job note.
• 2. Selects currency (16 options) and draft language (7 options).
• 3. Sends note to a local Ollama model (Qwen2.5‑1.5B).
• 4. Model returns structured JSON with client and line items.
• 5. App populates editable form; user can adjust rates, add tax, and review fields.
• 6. User prints or saves the finalized PDF.

Evidence

Talk2Invoice helps an independent service worker turn a short note about a finished job into a client-ready invoice draft.

Source

The person can type a note or use optional browser dictation, choose one of 16 currencies and one of seven draft languages… then ask an open-weight model to suggest the client and line items.

Source

The UI is plain JavaScript served by a small Node.js server. When a user requests a draft, the server calls an Ollama-compatible local model and requests structured JSON.

Source

The recommended model is Qwen2.5 1.5B, an open-weight, multilingual model available through Ollama.

Source

Talk2Invoice does not record or upload audio itself.

Source