Documentation
Everything you need to get started with VivaDicta and get the most out of it.
Getting Started
Transcription
How It Works
The 7-stage pipeline from speech to polished text — cleanup, formatting, AI, and paste.
Transcription Models
Overview of local and cloud transcription models — speed, accuracy, and privacy trade-offs.
Recommended Models
Our top picks for different use cases — free, accurate, private, or fast.
Speaker Labels
Prefix each turn with Speaker A, Speaker B - configured per Viva Mode, on the models that support it.
Transcribe Files
Queue up any number of audio or video files, pick a mode once, and transcribe them in a batch.
Cloud Providers
Groq Setup
Free forever cloud transcription — the fastest way to get started.
Cohere Setup
Best-in-class accuracy (5.4 WER) with free trial — no credit card required.
Deepgram Setup
$200 free credits for new accounts — high-accuracy transcription.
ElevenLabs Setup
Free tier with excellent multi-language support via Scribe 2.
Mistral Setup
Free Experiment plan with Voxtral transcription — open-source and multilingual.
Cerebras Setup
The fastest AI processing measured - $5 of credits, and the card can go straight back out.
Other Providers
Gemini, Mistral (Voxtral), Soniox, and OpenAI-compatible endpoints.
AI Processing
Enhance Any Text
Select text in any app, tap Right Option, and rewrite it in place - magic sequences, quick slots, and per-preset hotkeys.
AI Processing Overview
How VivaDicta uses AI to polish your transcriptions with built-in and custom presets.
AI Providers
Set up Claude, GPT, Gemini, Groq, Ollama, MLX, and more for AI text processing.
AI Model Rankings
We measured every AI model on cleaning up real dictation — quality, speed, and which ones to actually pick.
AI Assistant
Ask AI anything — answer questions, generate content, and work with screen context.
Translation
Speak in any language and get translated text — built-in or custom language presets.
Local AI Processing
Apple Intelligence, MLX models, and Ollama — free, private, on-device AI.
MLX Models
Qwen 3.5, GPT OSS, Ministral — run open-source LLMs locally on Apple Silicon.
AI Rewrite
Process any note with AI presets — generate summaries, key points, translations, and more as variations.
Use Your Subscriptions
Sign in with ChatGPT, Copilot, or Claude — use AI at no extra cost.
OAuth Sign-In Setup
Sign in with ChatGPT, Gemini, or GitHub Copilot — visual step-by-step guide.
CLI Agents Setup
Use your Claude, ChatGPT, or Gemini subscription for AI processing — no API keys needed.
VivAgents Server
Share CLI agents (Claude, Codex, Gemini) from your Mac with iOS devices.
Features
Viva Modes
Auto-configure transcription and AI per app or website.
Keyboard Shortcuts
Global shortcut, preset switching, and mini recorder controls.
Transcription History
Browse, search, play back, filter, and export your transcriptions.
Reminders & Calendar
Turn a note into Apple Reminders drafts and Calendar events, reviewed before anything is added.
Tags & Organization
Custom tags, source tracking, and filtering for your transcriptions.
Export Transcriptions
Export to CSV, JSON, or Markdown — individually or in bulk.
Vocabulary & Replacements
Custom vocabulary and automatic word replacements for domain-specific terms.
YouTube Transcription
Transcribe YouTube videos by pasting a URL.
iCloud Sync
Sync transcriptions, presets, and settings between Mac and iOS.
Automation & Integrations
MCP Server
Let Claude, Cursor, Codex, and other AI assistants search your history and trigger rewrites via the Model Context Protocol.
Command-Line Interface
One Homebrew install. Let Claude Code, Codex, or any AI agent drive transcription, search, and AI rewrites from their terminal.
CLI Skills for Agents
Drop-in skill pack that teaches Claude Code, Cursor, Codex, and other agents how to use the vivadicta CLI.