The SSE (Speech Signal Enhancement) is a 3rd party product by Cerence™.
The SSE is an advanced, modular library of field-proven algorithms specifically designed to improve the quality of voice communication. It enhances the performance of in-car hands-free communication and speech recognition systems. The library contains algorithms like beamforming, echo cancellation, noise reduction, wind noise removal, automatic gain control, bandwidth extension etc. (For more details see the SSE product specification.)
The library supports a large variety of configurations – from a simple single channel system to a multi-channel solution with many additional features. By means of SCD (SSE Configuration Data) the algorithms can be adjusted to the system and the acoustic environment they are used in. SSE gets a various number of microphone signals representing audio captured by microphones in the car cabin, by the IoT device or similar. For hands-free applications it may also get a receive input signal representing the signal from the far-end speaker. The optional reference signals are used by the echo canceller in hands-free applications and speech recognition with barge-in.