AbhishekBarali/SpeakoFlow

What it solves

SpeakoFlow is a cross-platform voice dictation and AI assistant tool that allows users to type with their voice in any application. It eliminates the need for manual typing by providing a system-wide voice-to-text interface, while also integrating an AI assistant for tasks like translation, explanation, and hands-free conversation.

How it works

The application uses a combination of local and cloud-based models for speech-to-text (STT), large language models (LLMs), and text-to-speech (TTS). Users trigger actions via keyboard shortcuts to dictate text, ask the AI assistant questions about selected text or the screen, or engage in spoken conversations. It supports a wide range of models including Whisper, Parakeet, and local engines like llama.cpp, as well as cloud providers like OpenAI and Anthropic.

Who it’s for

It is designed for people who want to increase their productivity by replacing typing with voice, users who need an AI assistant accessible across all their apps, and professionals who need to transcribe and summarize meetings automatically.

Highlights

• System-wide Dictation: Type in any app using voice with customizable keyboard shortcuts.
• AI Assistant: Ask questions, translate text, or replace selected text with AI-generated answers.
• Hands-free Conversation: Engage in spoken dialogues with the AI assistant using local or cloud voices.
• Meeting Transcription: Records microphone and system audio to create transcripts and structured meeting notes (summary, decisions, next steps).
• AI Cleanup: A specialized small model (SpeakoFlow Mini) that removes filler words and fixes grammar/punctuation in real-time.
• Privacy-First: Local transcription and processing by default, with no telemetry or accounts required.
• Flexible Model Support: Supports 65+ speech models and various LLM providers (local via Ollama/LM Studio or cloud API keys).