Skip to main content
Faseeh offers state-of-the-art Arabic voice synthesis models designed to handle various dialects and use cases.

​
Meet the Models

Faseeh v1 Preview

High-quality Arabic voice synthesis model
Natural-sounding Arabic speech
Multiple Arabic dialects supported
High-quality voice generation
Optimized for clarity and naturalness

Faseeh v1 Turbo Preview

Fast Arabic voice synthesis model
Low-latency generation
Multiple Arabic dialects supported
Faster model with optimized performance
Ideal for real-time applications

​
Model Comparison

FeatureFaseeh v1 PreviewFaseeh v1 Turbo Preview
QualityHighHigh
SpeedStandardFast
LatencyStandardLow
Use CaseGeneral purposeReal-time applications
DialectsMultiple Arabic dialectsMultiple Arabic dialects

​
Using Models

Models are used in the following endpoints:
  • Text-to-Speech Generation: POST /text-to-speech/:model_id (with streaming: false for WAV file)
  • Text-to-Speech Streaming: POST /text-to-speech/:model_id (with streaming: true for PCM16 stream)
  • WebSocket Streaming: WSS /text-to-speech (with model_id in initConnection)
Model Selection: Choose Faseeh v1 Preview for general use cases requiring high quality, or Faseeh v1 Turbo Preview for applications requiring low latency and faster generation.