ai-coustics Speech Enhancement SDK for Rust

[!WARNING] You must use a Rust version different from beta 1.92, which was used to build the static libraries. A solution is currently in development.

What is this SDK?

Our Speech Enhancement SDK delivers state-of-the-art audio processing capabilities, enabling you to enhance speech clarity and intelligibility in real-time.

Quick Start

Generate your SDK License Key

To use this SDK, you'll need to generate an SDK license key from our Development Portal.

Please note: The SDK license key is different from our cloud API product. If you have an API license key for our cloud services, it won't work with the SDK - you'll need to create a separate SDK license key in the portal.

Integration

Enable the download-lib feature to automatically download the library when building the crate.

[dependencies]
aic-sdk = { version = "0.11.0", features = ["download-lib"] }

If you want to provide your own library, use the AIC_LIB_PATH environment variable to specify the path to the directory where the library is located.

Example Usage

use aic_sdk::{EnhancementParameter, Model, ModelType};

fn main() -> Result<(), Box<dyn std::error::Error>> {
    let license_key = std::env::var("AIC_SDK_LICENSE")?;

    // Create a speech enhancement model by selecting a model type
    // and passing your license key as an &str
    let mut model = Model::new(ModelType::QuailS48, &license_key)?;

    // Initialize the model with your audio settings
    model.initialize(48000, 1, 480, true)?;

    let mut audio_buffer = vec![0.0f32; 480];

    // The process function is where the actual enhancement is happening
    // This is meant to be called in your real-time audio thread
    model.process_interleaved(&mut audio_buffer)?;

    // You can also adjust parameters during processing
    model.set_parameter(EnhancementParameter::EnhancementLevel, 0.8)?;

    // For planar audio processing (separate channel buffers)
    let mut audio = vec![vec![0.0f32; 480]; 2]; // 2 channels, 480 frames each
    let mut audio_refs: Vec<&mut [f32]> = audio.iter_mut().map(|ch| ch.as_mut_slice()).collect();
    model.initialize(48000, 2, 480, true)?;
    model.process_planar(&mut audio_refs)?;

    Ok(())
}

Running the Example

To run the example, make sure you have set your license key as an environment variable:

export AIC_SDK_LICENSE="your_license_key_here"

Then use the following commands to configure, build and run the example:

cargo run --example basic_usage --features download-lib

Support & Resources

Documentation

Basic Example - Sample code and integration patterns

Looking for Other Languages?

The ai-coustics Speech Enhancement SDK is available in multiple programming languages to fit your development needs:

Platform	Repository	Description
C	`aic-sdk-c`	Core C interface and foundation library
C++	`aic-sdk-cpp`	Modern C++ interface with RAII and type safety
Python	`aic-sdk-py`	Idiomatic Python interface
JavaScript/TypeScript	`aic-sdk-node`	Native bindings for Node.js applications
Web (WASM)	`aic-sdk-wasm`	WebAssembly build for browser applications

All SDKs provide the same core functionality with language-specific optimizations and idioms.

Get Help

Need assistance? We're here to support you:

Issues: GitHub Issues
Technical Support: info@ai-coustics.com

License

This Rust wrapper is distributed under the Apache 2.0 license, while the core C SDK is distributed under the proprietary AIC-SDK license.

Made with ❤️ by the ai-coustics team

aic-sdk 0.11.0