Open Source JavaScript Speech to Text Software

OpenWhispr is a free, open-source speech-to-text application designed for privacy-focused voice dictation, transcription, and AI-assisted productivity. It converts spoken words into text that can be inserted directly into applications using a global keyboard shortcut. Users can process speech locally with models such as Whisper and NVIDIA Parakeet or choose cloud-based transcription services. The application supports multilingual dictation, translation, and integration with AI assistants for voice-driven tasks. It also offers meeting transcription with speaker identification and searchable notes. Its local processing options keep recorded audio on the user's device without requiring cloud transmission. Available for Windows, macOS, and Linux, OpenWhispr combines voice input, intelligent automation, and configurable transcription workflows.

AppGrabbit is a Windows 10/11 app that downloads, transcribes and converts videos and enhances images with AI on your own computer. Paste a link from YouTube (videos, Shorts, playlists), TikTok, Instagram, Facebook, X, Pinterest, Twitch or SoundCloud and the file lands in your folder in the quality you pick, up to 4K, or straight as ProRes or DNxHR for editing. Turn any video or your own recording into a clean .txt with a prompt for ChatGPT, Claude or Gemini, transcribed on your PC with Whisper, nothing uploaded. New in 3.3: enlarge photos, logos and drawings x2 or x4 with AI on your graphics card, and turn a logo or drawing into an SVG vector for print, Cricut or Canva, with a free preview. No account, no ads. Free plan: 5 credits a month, 1 more every 48 hours when you run out. Pro: USD 8.99 a month; passes of 2 days (USD 2.99) or 7 days (USD 3.99), no renewal.

Voice Conference Manager uses VoiceXML and CCXML to control speech recognition, text to speech, and voice biometrics for a telephone conference service. Say the names or numbers of people and VCM places them into the call. Can be hosted on public servers

Transcrib — десктопное Electron-приложение для перевода аудио и видео в текстовый формат. Программа работает на базе мощной модели Whisper от OpenAI и Node.js. «Из коробки» в дистрибутив уже вшиты две легкие модели — `tiny` и `base`. Остальные (`small`, `medium`, `large`) можно скачать и обновлять прямо внутри приложения по мере необходимости. Логика простая: чем сложнее модель, тем выше качество текста, но тем дольше идет обработка.