Skip to main content

​
Overview

By default, Superwhisper delivers fast performance. However, if you’re experiencing slowness or want to improve the app’s responsiveness, the tips below point out the areas that tend to have the biggest impact on how fast the app runs. Several factors can affect Superwhisper’s performance:

System Resources

Your computer’s processing power, RAM, and available memory directly impact performance, especially when using local models.

Internet Connection

Cloud-based processing depends on your internet speed and stability for optimal performance.

Model Selection

Different AI models will vary in performance based on your specific use case.

Processing Type

Voice transcription and AI processing are separate operations with different performance implications.

​
Local vs Cloud Models

The choice between local and cloud models can significantly affects performance:

​
Local Models

Local models run offline, directly on your computer, which means that your system specifications directly influence processing speed.
1

Optimize Voice Model Active Duration

Superwhisper Voice Model Active DurationIn Superwhisper’s Advanced Settings, you can adjust how long voice models remain loaded:
  • Settings range from 10 seconds to 1 hour
  • Shorter durations free up system resources but require reloading for new dictations
  • Longer durations keep models ready but consume more system resources
2

Consider Your Hardware

For optimal local model performance:
  • Ensure adequate RAM
  • More powerful CPUs will process dictations faster
  • If your Mac struggles with local models because of its specifications, consider closing resource-intensive applications when using Superwhisper

​
Cloud Models

Cloud models process your dictations on remote servers:
1

Select the Appropriate Cloud Model

Superwhisper uses different models for voice transcription and AI processing. Some are better for certain tasks or types of content.
2

Monitor Your Internet Connection

Depending on your system specs, cloud models generally offer faster processing than local options, but require a stable internet connection.

​
Optimizing Processing Speed

1

Disable AI Processing When Not Needed

Transcription and AI processing are separate operations:
  • Using voice mode only (without AI processing) provides the fastest results
  • Consider this option when you don’t need special formatting
2

Choose the Right Model for Your Task

Match the model to your specific needs:
  • Short dictations: Ultra Cloud model
  • Local and fast: Parakeet model (English only)
  • Offline reliability & privacy: Local models
  • Longer dictations: Any model
3

Monitor Processing Times

Use the History menu to identify performance bottlenecks:
  • Check processing time for voice transcription
  • Check processing time for AI processing
  • Identify which part of the process is causing delays and switch model if necessary

AI Models

Learn how to set your preferred AI models in Superwhisper

History

Discover how to use the History panel for troubleshooting

Advanced Settings

Explore all available configuration options

Reporting Issues

Learn how to effectively report problems to the Superwhisper team
⌘I