Developer Home

System Components

Using Machine Learning, the Immersitech Engage™ Core platform is trained to remove over 52,000, 48 kHz sounds that take away from providing an engaging digital experience. From dogs barking to the constant hum of an air conditioner,  you’re covered. During AI model training, we perform a real-time combination of noise, clean speech, and impulse response data to create optimum coverage for noise cancellation. For special use cases, we can work with you to customize the model to remove specific types of noise.

Tightly coupled with Noise Cancellation, Voice Clarity enhances the quality of participant voices to ensure that all participants sound natural, clear, and crisp – not over-processed. We provide auto gain control to put loud and quiet talkers on a common level, and even equalize the tonal qualities of the audio stream to provide a more consistent and understandable voice character.

Using patented audio processing techniques, we can place each participant at distinct location within 3D space around the listener. The placement of participants in 3D space allows for a more natural, realistic experience that reduces digital meeting fatigue that many experience in traditional online meetings. Listeners can use headphones or a single pair of speakers to experience the 3D spatial effects.

Example: Processing audio for 3 person conference

• Participants can have multiple input, output, and metadata streams that are processed and mixed using low-latency methods to maintain a high-quality experience.
• Positional data gives Service Providers an opportunity to create immersive experiences by sending real-time positional data or defining targeted user positions depending on your desired use case.
• Developers and Service Providers can configure the Immersitech Engage™ Core in various combinations of the 3 primary features using a simplified set of APIs.

Client Side Implementation

• The SDK is deployed as part of your compiled application.
• Noise Cancellation and Voice Clarity can be deployed quickly without external dependency.
• 3D Spatial Audio can be enabled using real-time coordinate information to render sounds in specific locations.

Hybrid Implementation

• The SDK installed on the client-side handles Noise Cancellation and Voice Clarity.
• A central server-based solution handles 3D Spatial Audio.
• The processing power is split between the server and application, ensuring less strain on your participant’s machines.