Voice Input for NVDA

Overview

Voice Input is an NVDA add-on that provides ultra-fast (<0.5s), focus-independent speech-to-text dictation with smart spoken punctuation parsing and real-time multi-language translation directly into any active Windows application.

Key Accessibility Advantages:

  1. Focus-Preserving Direct Injection: Injects transcribed text directly into the focused application control via low-level SendInput, without stealing focus or disturbing NVDA review cursors.
  2. Screen-Reader-Centric Spoken Punctuation: Bilingual parser (English & Indonesian) translates spoken commands (e.g. koma, titik, tanda tanya, comma, new line) into proper symbols, complete with automatic capitalization and spacing formatting.
  3. Instant Translation: Speak in your native language and have translated text immediately typed into any chat or document.
  4. State-of-the-Art Whisper Engine: Powered by OpenAI Whisper Large v3 on Groq Cloud LPU hardware for sub-second recognition speed.
  5. User Key Management: Users configure their own free Groq Cloud API key in NVDA Settings. Zero hardcoded developer credentials exist in the source code.

Spoken Punctuation Reference

| Spoken Words (English / Indonesian) | Symbol Inserted | | :--- | :---: | | period / full stop / dot / titik | . | | comma / koma | , | | question mark / tanda tanya | ? | | exclamation mark / tanda seru | ! | | colon / titik dua | : | | semicolon / titik koma | ; | | open parenthesis / buka kurung | ( | | close parenthesis / tutup kurung | ) | | open quote / buka petik / kutip dua | " | | close quote / tutup petik | " | | single quote / petik satu / kutip satu | ' | | new line / baris baru / paragraf baru | \n | | slash / garis miring | / | | backslash / garis miring terbalik | \ | | underscore / garis bawah | _ | | hyphen / dash / minus / strip | - | | plus / tambah | + | | equals / sama dengan | = | | percent / persen | % | | asterisk / star / bintang | * | | at sign / at / keong | @ |

Keyboard Commands

You can customize this shortcut at any time via NVDA Menu -> Preferences -> Input Gestures -> Voice Input.

Getting Your Free Groq API Key (1-Minute Setup)

  1. Visit Groq Cloud Console.
  2. Sign in with your Google account.
  3. Click Create API Key, enter any name, and copy your key (starts with gsk_...).
  4. In NVDA, open Preferences -> Settings -> select Voice Input.
  5. Paste your API key into the Groq API key field and press OK.

System Requirements