🎧 We're here to listen

Sonora is designed for focused, short voice memos. The 3‑minute limit encourages concise recordings and helps performance, battery life, and storage.

Absolutely. Sonora supports background recording with Live Activities integration. You'll see a real-time recording indicator in your Dynamic Island and on the lock screen. Recording can continue even if you switch apps, lock your device, or receive notifications.

Sonora offers two transcription modes:

• Cloud Transcription: You get 10 minutes per day of cloud‑based transcription. Audio is sent securely for processing and deleted after completion.
• Local Transcription (On‑Device): Unlimited. Audio is processed on your iPhone and stays on your device.

Sonora supports transcription in many languages and dialects, including English, Spanish, French, German, Chinese (Mandarin & Cantonese), Japanese, Arabic, Portuguese, and Russian. The service can automatically detect the spoken language in most cases.

Transcription accuracy depends on several key factors:

• Audio Quality: Clear speech in quiet environments can improve results
• Speaking Pace: Normal, conversational speed is ideal for accurate transcription
• Language Support: Some languages may have higher accuracy rates than others
• Technical Terminology: Specialized vocabulary or jargon may be less accurately transcribed
• Accents & Dialects: Strong regional accents might affect transcription accuracy
• Background Noise: Ambient sounds can interfere with speech recognition

Currently, Sonora displays transcriptions as generated by our advanced AI service to maintain the integrity of the original analysis. If you need to make corrections, you can easily copy the text and edit it in your preferred text editor. We're actively considering adding in-app editing capabilities in future updates based on user feedback and requests.

Transcription time depends on connection, server load, and content. The process starts automatically after recording, and a progress indicator is shown.

Sonora can generate AI‑powered insights from transcripts:

• 📄 TLDR Summaries: Concise, intelligent summaries that capture the essence of your recording
• 🎯 Theme Extraction: Identifies key topics, themes, and concepts mentioned in your voice memo
• ✅ Action Items: Automatically detects tasks, to-dos, and actionable items from your speech
• 💡 Key Insights: Highlights important points and decisions
• 📊 Sentiment Analysis: Understanding the emotional tone and context of your thoughts
• 🏷️ Smart Tagging: Automatic categorization to help organize your voice library

Our AI analysis quality depends on several factors, and we're transparent about both strengths and limitations:

• Performance: Clear, structured speech with specific topics can yield better results
• Content Structure: Recordings with defined objectives or organized thoughts are analyzed most effectively
• Professional Context: Work-related discussions and decision-making content typically receive highly accurate analysis
• Action-Oriented Speech: Recordings that mention specific tasks, deadlines, or decisions are particularly well-suited for AI insights
• Continuous Improvement: Our AI models are regularly updated to improve accuracy and understanding

Results may vary for highly casual conversations, stream-of-consciousness thoughts, or highly technical content in specialized fields.

Absolutely. Sonora puts you in complete control of your AI analysis experience. The analysis is entirely optional and only runs when you specifically request it, giving you full autonomy over when and how your content is analyzed. You can choose to:

• Run analysis immediately after recording
• Review the transcription first, then decide on analysis
• Skip analysis entirely for personal or sensitive recordings
• Batch analyze multiple recordings at once
• Set preferences for automatic analysis based on content type