Hello there, my name is Gary Sims from Android Authority, and I've stopped by the Silver stand, who make DSP technology, to meet up with Aaron and learn a bit about them and what they do. Hi, nice to meet you, hey Gary, so yeah, CA is basically a DP provider, we provide IP just similar business model to Arm's uh only for GSP, so we have four product lines basically, one for image processing and object recognition, your networ stuff like that, one for audio and voice, sensing Technologies another one for cellular Technologies, and another one for connectivity like Bluetooth and Wi-Fi. So it all comes in the form of IP where you can integrate that into your smartphone or tablet and um here we have a solution which is actually integrated into the Galaxy S7 doing the speech recognition, so whenever you say "Okay Google" to your Galaxy S7, that's our technology inside, uh through one of our silicon Partners DSP group they make the chip and the interesting thing about the Galaxy S7 is that Samsung chose to use the Snapdragon a20 as the application processor now, the a20 has its own ultra low power island for this type of applications, the speech recognition etc for always on applications and despite that, Samsung still chose to use the DSP group solution with our DSP inside specifically because of its ultra low power consumption even comparing to that low power island of the Snapdragon. So once you look at the total PCB, you can see it here and that over there in blue is the DSP group chip doing the speech recognition, so from my perspective, that's a very important design win indicating of the ultralow power consumption of our solution, that's really amazing.

I didn't know that that was there inside the S7, that's amazing, and I see here also we have a smart speaker which is the kind of things that are coming to our home nowadays, we've got products from Google and from Amazon, you're also in this field. What do you do here? Yeah, so basically, it's similar concept of speech recognition only that in the smart home space and smart speakers, you want to have the far-field capability unlike smartphones, which are usually near-field, although we start hearing about the trend of integrating farfield capabilities even into smartphones now, why is that because you have four mics or three or four mics in most smartphones today, so you could theoretically do the far-field capabilities, the thing is, power consumption again the uh smartphone battery is much smaller compared to the smart speaker battery, so power consumption is crucial. Again, um but still again with our silicon part DSP group, we have multiple design wins already for smart speakers, and the difference here, in the Galaxy S7 for example, this is a single microphone solution in a smart speaker as you can see here in this uh demonstration, there are three microphones and the number ranges between two mics like in the Google home to seven mics like in the Amazon Echo or eight 8 mics like in the Ding Dong, but basically it's a multi-mic application and again power consumption is crucial. So that's a similar concept only applied to a multimicrophone setup, and it also involves additional Technologies like beamforming acoustic echo cancellation de-reverberation so other technologies needed for the FarField pickup to handle all those echoes and also this uh device emits music, so you have to consider the music that's why we have the acoustic echo cancellation, DSP is brilliant for all those different things, you just mentioned exactly that involves intensive signal processing, so and an MCU or CPU is a no-go for that type of application, it's not just like a sensor, you know like 100 Hz sensor, Those sensors those uh acoustic sensors they either 16 to all the way to 48 kHz so and there are multiple uh microphones in such a setup so we're talking about multiple sensors and the horsepower um which is required for processing that increases dramatically and yet staying low power which is the secret yeah exactly because battery life is key.

"WEBVTTKind: captionsLanguage: enhello there my name is Gary Sims from Android authority we're here at Mobile World Congress I've stopped by the Siver stand who make DSP technology and I've met up here with Aaron he'll tell me a bit about them and what they do hi nice to meet you hey Gary so uh yeah CA is basically a DP provider we provide IP just similar business model to Arms uh only for GSP so we have four product lines basically one for uh image processing and object recognition your networ stuff like that one for audio and voice and sensing Technologies another one for cellular Technologies and another one for connectivity like Bluetooth and Wi-Fi so it all comes in the form of Ip where you can integrate that into your so and um here we have a solution which is actually integrated into the Galaxy S7 doing the speech recognition so whenever you say Okay Google to your Galaxy S7 that's our technology inside uh through one of our silicon Partners DSP group they make the chip and the interesting thing about the Galaxy S7 is that Samsung chose to use the Snapdragon a20 as the application processor now the a20 has its own ultra low power Island for this type of applications the speech recognition etc for always on applications and despite that Samsung still chose to use the DSP group solution with our DSP inside specifically because of its ultra low power consumption even comparing to that low power island of the Snapdragon so once you look at the total PCB you can see it here and that over there in blue is the DSP group chip doing the speech recognition so from my perspective that's a very important design wi indicating of the ultr L power consumption of our Solution that's really amazing I didn't know that that was there inside the S7 that's that's amazing and I see here also we have a smart speaker which is the kind of things that are coming to our home nowadays we've got products from Google and from Amazon you're also in this field what what do you do here yeah so basically it's it's a similar concept of speech recognition only that in the smart home space and smart speakers you want to have the far field capability unlike smartphones so smartphones are usually near field although we start hearing about the trend of integrating farfield capabilities even into smartphones now why is that because you have four mics or three or four mics in most smartphones today so you could theoretically do the uh far field capabilities the thing is par consumption again the uh smartphone uh battery is much smaller compared to the smart speaker battery so power consumption is crucial again um but still again with the our silicon part DSP group we have multiple design WIS already for smart speakers and the difference here so in the Galaxy S7 for example this is a single microphone solution in a smart speaker as you can see here in this uh demonstration there are three microphones and the number ranges between two mics like in the Google home to seven mics like in the Amazon Echo or eight8 mics like in the ding-dong but basically it's a multi mic application and again par consumption is crucial so that's a similar concept only applied to a multim microphone setup and it h it also involves additional Technologies like beam forming acoustic Echo canellation de reverberation so other Technologies needed for the F Feld pickup to handle all those Echoes and also this uh device emits music so you have to cons the music that's why we have the acoustic Echo con the DSP is brilliant for all those different things you just mentioned exactly that involves intensive signal processing so and an MCU or CPU is a no go for that type of application it's not just like a sensor you know like 100 HZ sensor Those sensors those uh acoustic sensors they either 16 to all the way to 48 khz so and there's there are multiple uh microphones in such a setup so we're talking about multiple sensors and the horsepower um which is required for processing that increases dramatically and yet staying low power which is the secret yeah exactly because battery life is key yeah so now we've come over to the Bluetooth low energy part of your stand you've got an audio over B uh solution and of course B 5 has just come out and this is more low power and of course we're going to see this in our smartphones hopefully maybe next year our accessories are going to get this thing so it's a whole new thing and you're in that space right so we have uh we are the leading provider of IP for Bluetooth and Wi-Fi so we have a Bluetooth 5 solution and one of the main Innovations in Bluetooth 5 is as you mentioned the audio over Bluetooth low energy so until now we have audio over Bluetooth but that was very power consuming and now we want to have um much more much better power consumption so we have Bluetooth uh audio over blle Bluetooth low energy and consider for example in this application is the he AIDS or in general the hearables or the headset uh Market the wireless heads sets so the par consumption is critical because the batteries are very tiny especially in in application like a hearing aid so you don't want to mess with the batteries every few hours like you do with your Bluetooth headsets so par consumption is critical and indeed the audio over B enables you to enjoy audio over Bluetooth but with much lower power consumption than you could before Bluetooth 5 well this is amazing so you are like the hidden Heroes of the of the uh voice and of and of audio because your stuff is built into other people's chips and that's we as the consumers at the end benefit from that and I think that's absolutely great thank you very much for your time today thank you\n"