Skip to main content
As the ai-coustics SDK evolves, we occasionally deprecate older models in favor of newer, more efficient and performant ones. This guide provides an overview of the deprecated models and recommendations for migrating to the latest versions.

​
Quail

SDK, Real-time, Human-to-Machine With the release of the Quail Voice Focus 2.0 model, the older 1.0 and 1.1 VF models have been deprecated. We recommend migrating to the Quail VF 2.0 model for improved ASR performance and VAD capabilities.
  • ID: quail-vf-1.1-l-16khz
  • Last supported core SDK version: 0.14
  • File size: 35 MB
  • Window length: 10 ms
  • Optimal sample rate: 16 kHz
  • Optimal num frames: 160
  • Minimal algorithmic delay: 30 ms
  • ID: quail-vf-l-16khz
  • Last supported core SDK version: 0.14
  • File size: 35 MB
  • Window length: 10 ms
  • Optimal sample rate: 16 kHz
  • Optimal num frames: 160
  • Minimal algorithmic delay: 30 ms

​
Sparrow

SDK, Real-time, Human-to-Human Starting with core SDK 0.15, the smaller, ultra-low-latency Sparrow models have been deprecated. There are currently no plans to develop alternatives to these models. Sparrow model variants S and L continue to be supported and are recommended for real-time speech enhancement in human-to-human communication scenarios. If you have a need for ultra-low-latency speech enhancement models, please reach out to us to discuss your use case and requirements.
  • ID: sparrow-xs-48khz
  • Last supported core SDK version: 0.14
  • File size: 1.62 MB
  • Window length: 10 ms
  • Native sample rate: 48 kHz
  • Native num frames: 480
  • Minimal algorithmic delay: 10 ms
  • ID: sparrow-xxs-48khz
  • Last supported core SDK version: 0.14
  • File size: 1 MB
  • Window length: 10 ms
  • Native sample rate: 48 kHz
  • Native num frames: 480
  • Minimal algorithmic delay: 10 ms

​
Need help?