VoiceFlow Automate v2.0

A powerful Logseq plugin that transcribes voice notes and automatically creates tasks in Todoist using OpenAI's Whisper and GPT models.

🎙️ Important: AAC Audio Support

Logseq's mobile app records audio in AAC format by default. Since OpenAI's Whisper API doesn't support AAC files directly, you'll need to run the Audio Converter Server to process these files.

Audio Converter Server Setup

The logseq-audio-converter-server provides automatic AAC to M4A conversion for mobile recordings.

Quick Setup:

1. Install Prerequisites:

   • Python 3.8+
   • FFmpeg (Download)

2. Clone and Install:

   git clone https://github.com/GonzRon/logseq-audio-converter-server
   cd logseq-audio-converter-server
   pip install -r requirements.txt

3. Run the Server:

   python aac_converter_server.py

   The server runs on http://127.0.0.1:3456 by default.

4. Configure in VoiceFlow Settings:

   • AAC Converter Host: 127.0.0.1
   • AAC Converter Port: 3456

Note: Desktop users recording in MP3/M4A/WAV formats don't need the converter server.

✨ Key Features

• Voice Transcription: Convert audio files to text using OpenAI's Whisper API
• AI Summarization: Generate intelligent summaries and extract actionable tasks
• Direct Todoist Integration: Create tasks without external dependencies
• Smart Tag Detection: Automatically trigger actions using spoken hashtags
• Hierarchical Tasks: Create master tasks with subtasks for complex projects
• Due Date Recognition: Natural language processing for deadlines
• Mobile Support: Full AAC audio support with converter server

🚀 Quick Start

1. Installation

1. Download the plugin from the Logseq marketplace (or clone from GitHub)
2. Enable the plugin in Logseq settings

2. Configuration

Required: OpenAI Setup

1. Get your API key from OpenAI Platform
2. Open plugin settings in Logseq (⚙️ → Plugins → VoiceFlow Automate)
3. Enter your OpenAI API key
4. Select your preferred model (GPT-4o recommended)
5. Test the connection

Optional: Todoist Integration

1. Get your API token from Todoist Settings
2. Enter the token in plugin settings
3. Test the connection
4. Configure project mappings (optional)

For Mobile Users: Audio Converter

1. Follow the Audio Converter Server Setup above
2. Configure converter settings in plugin
3. Test with a sample AAC recording

3. Basic Usage

1. Record Audio:

   • Mobile: Use Logseq's built-in recorder (creates AAC files)
   • Desktop: Use any audio recorder plugin (MP3/M4A/WAV)

2. Process Voice Note:

   • Place cursor on the block with audio file
   • Use slash command /voiceflow or keyboard shortcut Cmd/Ctrl+Shift+V

3. Automatic Processing:

   • Transcribes the audio
   • Creates AI summary (if enabled)
   • Extracts and creates tasks in Todoist (if #todo spoken)
   • Generates a new Logseq page with everything

🎯 Spoken Hashtag Commands

Use these spoken hashtags in your voice notes to trigger actions:

Todo Creation

• "hashtag todo" or "hashtag task" - Create task(s) in Todoist
• "hashtag to-do" - Alternative spelling supported

Processing Mode

• "hashtag ai" - Use AI summarization (overrides default)
• "hashtag direct" or "hashtag literal" - Use literal transcription

Project Routing

• "hashtag [project-tag]" - Route to specific Todoist project
• Example: "hashtag work" routes to your Work project

Priority Setting

• "hashtag urgent" or "hashtag high" - Set priority 4
• "hashtag medium" - Set priority 3
• "hashtag low" - Set priority 1

Due Date Setting

• "due date tomorrow" - Sets due date to tomorrow
• "due date next Friday" - Natural language date parsing
• "due date January 15th" - Specific dates supported

📋 Examples

Example 1: Simple Task

"Remember to call the dentist tomorrow, hashtag todo"

Creates: Single task "call the dentist" with due date tomorrow

Example 2: AI-Processed Project Tasks

"Plan the office party: book venue, order catering, send invitations, 
arrange decorations, hashtag todo hashtag ai hashtag work"

Creates:

• Master task: "Plan Office Party" (in Work project)
• Subtasks: Book venue, Order catering, Send invitations, Arrange decorations

Example 3: Urgent Task with Due Date

"Finish the quarterly report, due date Friday, hashtag todo hashtag urgent"

Creates: High-priority task with Friday deadline

⚙️ Advanced Configuration

Project Mappings

Map spoken hashtags to Todoist projects in the UI:

1. Click "Load Projects for Mapping" button
2. Edit the generated JSON to customize tags:

{
  "#work": {"id": "2234567890", "name": "Work"},
  "#personal": {"id": "2234567891", "name": "Personal"},
  "#health": {"id": "2234567892", "name": "Health"}
}

Custom AI Prompts

Customize how AI processes your voice notes by configuring custom prompts in settings.

Transcription Modes

• Literal: Exact transcription → single task
• AI: Intelligent processing → multiple structured tasks

🔧 Troubleshooting

Audio File Issues

AAC Files (Mobile)

• Ensure Audio Converter Server is running
• Check converter host/port settings match server
• Verify FFmpeg is installed correctly
• Server logs: Check aac_converter.log

Supported Formats

• With Converter: AAC, MP3, M4A, WAV, WebM
• Without Converter: MP3, M4A, WAV, WebM (no AAC)

Todoist Tasks Not Created

1. Verify API token is correct
2. Test connection in settings
3. Check project mappings are valid
4. Ensure you spoke a trigger tag (#todo)
5. Check browser console for errors

OpenAI Errors

• 401: Invalid API key
• 429: Rate limit or quota exceeded
• 400: Check audio file format
• Network Error: Check internet connection

Converter Server Issues

• Connection Refused: Start the converter server
• Port Already in Use: Change port in both server and plugin settings
• FFmpeg Not Found: Install FFmpeg and check PATH

📝 Workflow Tips

Mobile Workflow

1. Record voice note in Logseq mobile
2. Ensure converter server is running on your computer
3. Sync to desktop
4. Process with VoiceFlow

Desktop Workflow

1. Record with audio plugin (MP3/M4A format)
2. Process immediately with VoiceFlow
3. No converter needed for non-AAC files

Best Practices

• Quick Capture: Record → Add #todo → Auto-process
• Batch Processing: Record multiple items → Use AI mode for smart extraction
• Project Organization: Set up project mappings for automatic routing
• Meeting Notes: Record meeting → AI summarizes → Tasks created automatically

🆕 What's New in v2.0

• Direct Todoist Integration: No external plugins required
• Enhanced Tag Detection: Better recognition of spoken hashtags
• Hierarchical Task Support: Create structured task lists
• Due Date Recognition: Natural language date parsing
• AAC Support: Full mobile audio support with converter server
• Improved AI Processing: Better task extraction with GPT-4

📊 Performance Notes

• Transcription Speed: 10-30 seconds for typical voice notes
• AAC Conversion: 1-3 seconds via converter server
• AI Processing: 2-5 seconds for summary and task extraction
• File Size Limits: 25MB for audio files (configurable in converter)

🔒 Privacy & Security

• Local Processing: AAC conversion happens on your machine
• API Security: Keys stored locally in Logseq settings
• No Data Storage: Converter server doesn't store files
• Temporary Files: Auto-cleaned after processing

📄 License

MIT License - See LICENSE file for details

🤝 Contributing

Contributions welcome! Please submit issues and pull requests on GitHub.

Related Projects

• logseq-audio-converter-server - AAC conversion service
• Logseq - Privacy-first knowledge base

🙏 Credits

• OpenAI for Whisper and GPT APIs
• Todoist for their excellent API
• FFmpeg for audio conversion capabilities
• Logseq community for feedback and support

📞 Support

For issues:

1. Plugin Issues: Open issue on GitHub
2. Converter Issues: Check converter repo
3. API Issues: Check OpenAI/Todoist documentation

Debug Mode

Enable Debug Mode in settings for detailed logging in browser console (Ctrl+Shift+I).