Open Source Voice AI Apps
Explore open source voice AI tools for speech-to-text dictation, text-to-speech, voice cloning and voice assistants, plus the commercial products they replace.
Apps in Voice AI
Language
License
OpenWhispr
AI Assistants · Productivity · Voice AI
Privacy-first, cross-platform voice-to-text with local AI and cloud options
Epicenter
Developer Tools · Knowledge Management · Note Taking
A local-first monorepo led by Whispering, an open-source speech-to-text app, built on an MIT toolkit that turns your data into plain Markdown and SQLite files you own instead of a database you rent.
Handy
Productivity · Voice AI
Free, offline, open-source speech-to-text that pastes directly into any app on Windows, macOS, and Linux.
VoiceInk
AI Assistants · Productivity · Voice AI
A native macOS dictation app that turns speech into text on-device with Whisper or Parakeet models, then cleans the result up with an AI pass tuned to whichever app you are typing into.
VoiceStudio
Mcp · Music Audio · Voice AI
Open-source, fully local ElevenLabs alternative for voice cloning, voice design, video dubbing, dictation, transcription and audiobooks, with a local API and MCP server for agents.
Amical
AI Assistants · Note Taking · Voice AI
Local-first AI dictation that understands your active app — private, offline, and built for speed.
Voicebox
AI Development · Productivity · Voice AI
Clone voices, dictate anywhere, and give AI agents your voice — all locally.
fish-speech
AI Development · Developer Tools · Music Audio
SOTA open-source dual-autoregressive text-to-speech model with rapid voice cloning, inline emotion tags, and real-time streaming inference across 80+ languages.
Rasa Open Source
AI Assistants · AI Development · Voice AI
Rasa Open Source is a Python machine learning framework for building contextual, multi-turn chatbots and voice assistants that understand natural language and maintain conversation state.
Frog
AI Assistants · Productivity · Voice AI
An open-source, on-device Mac desktop pet that types what you say, takes meeting notes with speaker labels, and talks back — a private alternative to Wispr Flow and Granola.
Ghost Pepper
AI Assistants · Voice AI
A 100% private, on-device macOS dictation and meeting-transcription app — hold Control to talk, transcribes and pastes locally via Apple Silicon with no cloud APIs or data leaving your Mac.
About Voice AI
Voice as an interface
Speech models got good and small at the same time. Whisper, Parakeet and a wave of open text-to-speech models now run on a laptop, which means voice software no longer has to send your audio to someone else’s server. This category collects the open source projects built on that shift, alongside the commercial products they compete with.
What’s in this category
- Dictation and speech-to-text — system-wide apps that let you talk into any text field and get clean, punctuated writing back. Many add an AI pass that removes filler words and adapts formatting to the app you’re typing in.
- Text-to-speech and voice cloning — models and studios that generate natural speech from text, clone a voice from a short sample, and stream audio in real time for narration, voiceovers or agents.
- Voice assistants — frameworks for building assistants that understand spoken requests, keep track of a conversation and act on it.
How to choose
Local or cloud. On-device tools keep audio private and cost nothing per use, but quality and speed depend on your hardware and the model you pick. Cloud services usually win on polish and cross-device sync, at the cost of sending recordings to a third party.
Model choice. Check which speech models a tool supports and whether you can swap them. Being able to move from a small, fast model to a larger, more accurate one matters more over time than any single feature.
Platforms. Many dictation apps are macOS-only. If you work across Windows, Linux or mobile, filter for that first.
Latency. For dictation and live assistants, the delay between speaking and seeing or hearing a result decides whether a tool feels usable. Streaming support is worth looking for.
Licensing for voices. For text-to-speech and cloning, read the model license as well as the code license — some open weights restrict commercial use.
