Back to Help Center
Getting Started

Understanding voice transcription

Last updated: February 8, 2024

Understanding voice transcription

Scribbes uses advanced AI to convert your speech into text with high accuracy.

How It Works

Our transcription engine:

  1. Captures your audio input
  2. Processes it through Deepgram's AI models
  3. Returns formatted text in seconds
  4. Learns from context to improve accuracy

Supported Languages

Currently supported:

  • English (US, UK, Australian)
  • Spanish
  • French
  • German
  • Portuguese
  • And 20+ more languages

Accuracy Tips

For best results:

  • Use a quality microphone
  • Speak in a quiet environment
  • Speak at a natural pace
  • Pronounce words clearly
  • Use punctuation commands ("period", "comma", "new paragraph")

Processing Time

  • Short recordings (< 5 min): 5-10 seconds
  • Medium recordings (5-15 min): 15-30 seconds
  • Long recordings (15+ min): 30-60 seconds

What Happens to Your Audio?

Your privacy is important:

  • Audio is processed securely
  • Not stored permanently
  • Never used for training
  • Deleted after transcription

Was this article helpful?

Can't find what you're looking for? We're here to help.