Frequently Asked Questions

Answers to common questions about VoiceWeaver's features and performance.

General Features

Q: Which file formats are supported?
A: VoiceWeaver supports major audio formats (mp3, wav, m4a, flac) and video formats (mp4, mov, mpeg4).

Q: Does transcription require an internet connection?
A: No. VoiceWeaver uses a local Whisper model for transcription. All data processing stays on your device for maximum privacy.

Q: How do the AI Summary and Assistant work?
A: The app features an MLX-powered local Large Language Model (e.g., Qwen2.5). It processes your transcripts locally to provide summaries and answer questions.

Q: How can I export subtitles?
A: After transcription, click the export button in the toolbar to save as SRT (SubRip), VTT (WebVTT), or plain text.

Troubleshooting

Q: Why does the status show "Model Missing"?
A: This means the transcription or AI models were not detected. Please ensure the required model folders are in the app's resource directory.

Q: Why is transcription slow?
A: Performance depends on your hardware. While functional on Intel-based Macs, performance is significantly better on Apple Silicon chips.