SpeakoFlow — Free, Open-Source Voice Assistant

A free, open-source voice assistant for Windows, macOS, and Linux.

1. Voice Dictation

• Dictate in any app: Speak and have your words typed wherever your cursor is, from emails and chats to documents and code editors.

• Live transcription: See text appear as you speak.

2. AI Cleanup

• SpeakoFlow Mini: A small model trained specifically for dictation.

• Clean text: Automatically removes filler words and fixes grammar and punctuation.

• Voice editing: Follow spoken commands like "scratch that".

• Writing styles: Choose Professional, Friendly, Concise, Formal, Casual, or create your own.

3. Ask Out Loud

• Select text in any app and tell the assistant what to do.

• Translate text, draft replies, or explain selected content.

• Copy the answer, insert it at your cursor, or replace your selection.

4. Voice Conversations

• Have hands-free conversations with your AI assistant.

• Hear responses spoken aloud.

• Interrupt the assistant while it is speaking.

5. Meeting Notes (Beta)

• Record calls without a bot joining the meeting.

• Transcribe your side and the other participants separately.

• Generate summaries, decisions, and next steps.

• Ask questions about your meetings afterwards.

6. Screen Vision, Web Search & Reminders

• Screen vision: Ask questions about errors, charts, or anything on your screen.

• Web search: Find information using your voice.

• Reminders: Set reminders by speaking.

• Screen vision and web search remain disabled until you enable them.

7. Private by Default

• Voice transcription runs locally unless you choose a cloud service.

• No telemetry or analytics.

• API keys are stored in your system keychain.

8. Your Choice of Models

• 65 local speech models, including Parakeet for English, Nemotron for 28 languages, and Whisper for 99 languages.

• Cloud speech services: ElevenLabs and Deepgram support live transcription.

• AI models: Use the built-in model, Ollama, LM Studio, or a cloud provider with your own API key.

• Local text-to-speech: Kokoro, Kitten, Pocket TTS, and Supertonic.

• Cloud voices: Use ElevenLabs for spoken responses.

9. Profiles & Memory

• Create different assistant personas for different tasks.

• Let the assistant remember your preferences and how you work.

• Memory is off by default, stored locally, and fully editable or erasable.

10. Usage Insights

• Track how many words you've dictated.

• Monitor your speaking speed.

• See how much time you've saved compared with typing.

11. Automatic Updates

• Install new versions directly from within the app.

12. Cross-Platform

• Available for Windows, macOS, and Linux.

13. Free & Open Source

• MIT licensed and completely free.

• No paid plans.

• No word limits.