VOCAL’s Space Application Development Board (SADB-C6727B) enables developers to implement algorithms for full-duplex hands-free voice communication systems. The processor hardware is similar to what would be deployed for spacecraft in a fully radiation hardened (rad-hard) design.
Spacesuit helmets with integrated loudspeaker and microphones provide astronauts the mobility and flexibility needed for modern space missions. However, the resulting acoustic coupling between the loudspeakers and microphones makes acoustic echo cancellation (AEC) software a requirement. The AEC requirement is also true of intercom systems of space stations.
VOCAL’s AEC software is an efficient fixed-point implementation for rad-hard processors, such as the TI C6727B DSP. The adaptive filter of the AEC software learns the echo impulse response of the system and cleanly removes the acoustic reflections of the loudspeaker from the microphone signal. This allows for full duplex voice communication between astronauts. The AEC quickly adapts to the changing acoustic environment. For example, when astronauts remove their helmet, the algorithm will learn the new echo response and continue to remove echoes.
In addition to AEC, supplemental voice quality enhancement algorithms can be applied to improve the intelligibility of the communication. VOCAL’s noise reduction algorithm is very effective at removing quasi-stationary noise sources, such as fan and instrument noise. Parametric Equalizers and Automatic Gain Control can be applied to further enhance the important speech components for clear voice communications.
SADB provides multiple audio input and output channels for hardware and software development and testing of the hands-free voice communication system. Audio inputs are handled by a TI ADS1278 high speed multichannel analog-to-digital converter (ADC) which has a variant qualified for space operation. Audio outputs may be generated from digital signal Pulse Density Modulation (PDM) or Pulse Width Modulation (PWM) signals or a commercial grade AIC3106 audio codec is provided for algorithm development and for comparison of audio performance to PDM/PWM generated audio signals. Digital audio inputs/outputs can also be connected to other processors (using the McASP signals directly).
Please contact us to learn more about VOCAL’s Space Application Development Board and Voice processing software.