What happens when advances in audio DSP combine with the increasing ubiquity
of loudspeakers and microphones.
Advances in signal processing and a growing abundance of loudspeakers and
microphones in everyday devices means there has never been a more interesting
time to be in the business of making air molecules vibrate in interesting and
Audio focussed software, and its associated hardware, are combining to foster
an amazingly fertile time for sound based technologies.
Signal processing for audio has a long and successful history of major
advancements. From Morse-code to DTMF, the dial-up modem, the mp3 file format — the echo cancellation that attenuates the sound of your own voice on a Skype call.
The innovation continues: pushing the envelop of what is possible using sound
through digital manipulation, are companies creating the sensation of touch
in midair using ultrasound, providing immersive VR experiences by digitally
modelling acoustic and psychoacoustic phenomena, and listening for
illegal logging in the amazon using machine learning running on smartphones.
The capabilities and value that sound based technologies can deliver via new
methods of signal processing continues to expand.
Chirp’s own technology has consistently pushed forward the state of the art
for reliable data transmission via sound. In the last year we have increased
our data rate in a wide array of acoustic conditions by 8X whilst maintaining
our industry leading reliability.
Through software the decades-old loudspeaker and microphone continue to be
given new superpowers.
As well as new capabilities, we also see the fundamental infrastructure needed
to create, control and analyse sound becoming ever more ubiquitous and
capable. At some point sound based technologies always have to interact with
the real world though a transducer: a loudspeaker or a microphone.
We need three things to manipulate, listen for, and understand sound: A
loudspeaker, a microphone and a processor. With each of these components
capabilities continue to expand, driven by the combination of material
advances, faster more efficient processors. More recent innovations in each of
these spaces include zero power microphones, ultrasonic loudspeakers used for physical levitation of objects, and ultra low power processing units.
When combined with DSP software, hardware’s sophistication and raw computing capability is unlocking new possibilities. For example, what happens when you combine 64 microphones with a super efficient custom processor? You get an incredibly accurate, but still battery operated and handheld 'sound
camera’ capable of detecting the precise location of any sound source in the environment around you.
Then there is ubiquity of the hardware components themselves. Over the last
decades the advent of the smartphone combined these three items and placed
them in everyones pocket — over the next decade voice assistants will make
them truly ubiquitous within our everyday environments.
The history of interacting with computers is an effort to seek increasingly
natural means to communicate and interact with our machines. Beginning with
punch-cards, through to the cryptic command line, to the graphical user
interface, the mouse and then the direct manipulation of on screen objects via
the touchscreen — the history of interacting with these devices has been a
slow but steady swing away from talking to them on their terms, towards a
arrangement where we can talk to them on our terms.
This is in part why we believe that voice based user interaction is a not
momentary trend — it is here not only to stay, but to become a constant
presence to the point of being taken for granted in our domestic and urban
environments. The strangeness once felt listening to a lone, early adopter,
mobile phone user walk down the street taking a personal call will give way to
general public acceptance and then of course reliance on the benefits that
these new technologies bring. It will become a social norm around being able
to access information and control your environment by speaking to it.
We believe voice interaction with devices will be everywhere. Their increasing ubiquity will usher in a set of opportunities by rolling out the hardware infrastructure that is a prerequisite for software audio technologies such as Chirp.
New software processes, more capable hardware and audio infrastructure that is
evermore present in our environment.
Chirp is very pleased to be in a field with such an abundance of innovation
and opportunity. The cooperation of hardware and software continues a long
history of progress within audio. Our team is motivated every day to push
forward the state of the art for data over sound and do our part to add to the
innovation within an incredibly exciting, innovative wider field.