Products & Technologies

Malaspina Labs develops real-time software algorithms which enhance speech comprehension & voice interfaces in noisy environments. Active algorithm research includes discrimination of speech-of-interest from high level background noise including competing background speech, real-time synthetic reconstruction of human speech, and the efficient encoding of speech for storage and digital communication.

Implementations of Malaspina Labs algorithms are offered individually or in combination as the VoiceBoost™ suite of speech enhancement solutions for ultra-low power devices. The VoiceBoost suite of products offer solutions to a number of real-time audio signal processing problems involving detecting and isolating a primary speech-of-interest from background noise within audio signals. VoiceBoost solutions are available for noise reduction, speech recognition assist, and voice command activation and interaction.

VoiceBoost solutions are compact and CPU-efficient implementations offered in binary form for supported DSP and ARM architectures or as fixed-point C-language source code with complete architecture, porting, and functional source code documentation.

VoiceBoost™ Real Time Speech Discrimination in Noise

The VoiceBoost Real Time Speech Discrimination solution detects and isolates speech from complex background noise within audio signals using a patent-pending mathematical model of human speech. The VoiceBoost Speech Discrimination solution easily integrates into existing audio processing pipelines and can be configured for either single or dual-core processing. The solution supports mixed audio input signals from single omni-directional or directional microphones, as well as spatially separated input signals from multiple microphones (beamforming). Read More

VoiceBoost™ Embedded Command and Control

The VoiceBoost Speech Command and Control solution is an automatic speech recognition solution with ultra-small memory footprint and minimal power requirements, designed specifically to address the needs of low-cost mobile, wearable and IoT devices. The solution executes in real-time entirely on the device’s DSP, microcontroller or SOC, allowing manufacturers to add an embedded speech interface to low cost devices without the capability or power consumption drawbacks of connecting to a cloud-based or host application. Read More

VoiceBoost™ Voice Activation

The VoiceBoost Speaker Independent Voice Activation solution and the VoiceBoost Speaker Dependent Voice Activation solution both offer keyword spotting with ultra-low power continuous listening capability for supported audio DSPs. The VoiceBoost Speaker Dependent Voice Activation solution also provides two-factor speaker verification using user-defined activation commands. Read More