Local Models
FloWords includes powerful AI models that run entirely on your Mac. No internet connection required, no data sent to external servers.
Why Local Models?
Section titled “Why Local Models?”- 100% Privacy - Your voice never leaves your device
- No Internet Required - Works offline, anywhere
- No Subscription - Use unlimited, no per-minute fees
- Fast Response - No network latency
Available Models
Section titled “Available Models”FloWords includes three local transcription engines. All of them run on your Mac and are multilingual.
| Model | Engine | RAM | Speed | Accuracy | Best For |
|---|---|---|---|---|---|
| Whisper Turbo (default) | OpenAI • Q5_0 (~547 MB) | ~2 GB | ~200-800 ms | ~7-8% WER | General use, best balance |
| Parakeet V3 | NVIDIA • INT8 (~640 MB) | ~2 GB | ~50-200 ms | ~6.34% WER | Maximum speed, low latency |
| Apple Speech | Native macOS (no download) | Minimal | ~100-500 ms | ~8% WER | Quick drafts, no download |
Downloading Models
Section titled “Downloading Models”- Open the Intelligence section in FloWords
- Find the Best Models Available section
- Pick the model you want (Whisper Turbo, Parakeet V3, or Apple Speech)
- Click Download if it isn’t installed yet
- Select it as your active transcription engine
Model Performance by Mac
Section titled “Model Performance by Mac”Apple Silicon (Recommended)
Section titled “Apple Silicon (Recommended)”- Hardware acceleration - models take advantage of the Apple chip
- Performance - ~2-3x faster than Intel
- Efficiency - minimal battery impact
- Recommendation - all three models run great
Intel Macs
Section titled “Intel Macs”- Performance - slower than Apple Silicon
- CPU usage - higher (the fan may spin up)
- Recommendation - Parakeet V3 or Apple Speech for better speed
Model Configuration
Section titled “Model Configuration”Language Settings
Section titled “Language Settings”- Auto-Detect - the model automatically detects the language
- Specify Language - set a specific language for faster, more accurate results
To set your language:
- Go to Settings > Model
- Find Language dropdown
- Select your preferred language
- Or choose Auto for detection
Memory Management
Section titled “Memory Management”FloWords monitors your system memory and warns you if you’re running low.
Low Memory Warnings
Section titled “Low Memory Warnings”If you see memory warnings:
- Switch to a lighter model - Apple Speech uses no download and minimal memory
- Close other applications - Free up system memory
- Restart FloWords - Releases cached memory
Model Storage
Section titled “Model Storage”Downloaded Whisper and Parakeet models are stored in:
~/Library/Application Support/com.prakashjoshipax.VoiceInk/models/Apple Speech is not downloaded — it’s part of macOS.
Troubleshooting
Section titled “Troubleshooting”Model Download Fails
Section titled “Model Download Fails”- Check your internet connection
- Ensure you have enough disk space
- Try Apple Speech (no download required)
- Restart FloWords and try again
Transcription is Slow
Section titled “Transcription is Slow”- Switch to a faster model (Whisper Turbo → Parakeet V3)
- Close other applications
- Check Activity Monitor for memory pressure
- Ensure your Mac isn’t thermally throttled
Poor Accuracy
Section titled “Poor Accuracy”- Switch to Whisper Turbo for higher accuracy
- Set the correct language instead of auto-detect
- Reduce background noise
- Speak more clearly and at a steady pace
- Add custom terms to your Dictionary
Model Won’t Load
Section titled “Model Won’t Load”- Restart FloWords
- Re-download the model
- Check console logs for errors
- Free up system memory
Next Steps
Section titled “Next Steps”- Configure AI Enhancement to polish your transcriptions
- Set up Cloud Providers as a backup option
- Add Custom Dictionary for specialized terminology