SOUND
ADVICE
www . professional-sound . com / soundadvice
MORE
AUDIO DSP & THE BLENDED LEARNING MODEL
By Shaun Robinson
While classes are in session , it remains unfeasible in many regions to send everyone back to campus at once , whether at K-12 schools or college and university campuses . As such , blended learning models have become a popular choice . With this choice comes a new set of challenges for reaching students on campus and at home in equal measure , and without diminishing returns . The good news is that AV technology continues to mature , and the interoperability between presentation systems and conferencing software continues to tighten .
Inside the classroom , video content is typically moved around using switching and routing devices powered over HDBaseT or IP . The latter provides a more direct avenue to the conferencing software from Zoom , Teams , or Skype , for example . These IP-based distribution systems will also efficiently accommodate audio moving on and off the school ’ s IP network .
While video is a critical component , the audio piece cannot be undervalued . There is substantial power to visualizing the lesson , yet the learning suffers if the audio quality is unreliable . This is where the value of the digital signal processor , or DSP , comes into vision .
Consider a hybrid learning environment on a university campus with 50 students in the classroom , and another 50 at home . In this scenario , 100 students are connected to conferencing software on their laptops , all of which are in communication . These laptops all connect from varied acoustical environments . This not only requires a platform to manage these audio signals as they move between sources and destinations – there needs to be a way to eliminate feedback , echo , and artifacts to enable clear two-way communication . Finally , there needs to be a way to control how these signals are being processed , and manage where these signals are moving to , be it a display , a projector , an amplifier , or a recording device .
The open architecture DSP is the clever device that brings the communication of all these devices together for the hybrid learning model .
NETWORKED EFFICIENCY Let ’ s first look at a networked video distribution , or AV-over-IP platform , that accommodates video routing through the classroom or lecture hall . AV-over-IP platforms from companies like Atlona and Crestron will carry audio with the video , but work in parallel with a DSP to process and deliver lecture audio to end points .
In an AV-over-IP ecosystem , that audio travels to the DSP in the digital domain , often with Dante networking acting as an off ramp . The presence of Dante is also useful for the broader AV-over-IP ecosystem , including the amplifiers and speakers serving the end points . The Dante-enabled DSP acts as the core audio device in the ecosystem , managing audio signals – and often , as in the case of Xilica , audio-related control functions – moving over IP through the space . On the input side , that is not only the extraction of audio from the AV-over-IP platform – that also means taking in a variety of microphones . This covers everything from simple gooseneck and lavalier microphones , to more complex arrays .
Understanding how to position these systems in a socially-distanced classroom is paramount . One concept is to provide each student with a compact base station , with each station networked and moving signals to the DSP . The DSP optimizes these signals before outputting them over the network to both the in-room speakers and the conferencing software . It is a workable concept that remains in use .
With hybrid learning on the rise , we are thankfully seeing emerging concepts and technology evolutions that offer greater returns . This includes ceiling microphone arrays , such as Sennheiser ’ s TeamConnect beam forming systems , which automatically scan the room for the loudest source . In this scenario , the architecture will use far fewer microphones to cover a much larger space – approximately 550 sq . ft . per mic . Now , the student and lecturer can speak without activating a base station and handling a microphone .
While a DSP is useful in either scenario , it is particularly beneficial for processing ceiling arrays . An intelligent DSP will efficiently manage the crossover of sound between these different microphones , which creates an intelligible audio experience for remote learners . The modern DSP can manage the heft of audio moving through it , cleanly process the audio , and move it back to the network to its final destination . And , it can do so without any adverse effects that originated from activity within the classroom .
CRISP & CLEAR Often , echo is the most prominent concern in hybrid learning environments . This is where the processing power of the DSP becomes its highest priority . Acoustic echo cancellation ( AEC ) eliminates the far-end audio as it moves through the DSP and into the conferencing platform . The removal of that far-end audio , in theory , removes any echo from the live audio presentation .
The DSP should provide the same efficiency from the opposite direction . Consider one of the remote students is speaking to everyone else , and is cursed with an unpleasant echo . That effectively halts any effective communication . AEC seeks to solve this problem , and is best applied when built into the DSP at the core of the UC platform .
A bad echo algorithm will also exhibit a “ double talk ” effect . In this example , with two students talking over each other , will cancel one completely . While this provides immediate relief , it eliminates interaction . A quality DSP will cope with multiple speakers without fighting itself . At the same time , it will cancel the echo and related issues originating on the front end .
Beyond AEC , the DSP will require the appropriate filters to EQ and process the room to optimize the audio output . This will result in a high-quality and pleasant audio reproduction over the in-room speaker system .
As more of these technologies move to the network , these levels also become easier to control over the UC platform . We ’ re entering a new era of hybrid learning , and thankfully the AV industry is stepping up with effective and efficient , fully-networked UC solutions at just the right time .
Shaun Robinson is the Vice President of Product Management for Xilica .
56 PROFESSIONAL SOUND