DingoBorat
Slim
Then they can't comment..
BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
Hi Dippy,
I haven't watched the video, but my records show that, in 2019, Synaptics were using a software NN:
US11076080B2 Under-display image sensor for eye tracking 2019-12-05
A method for eye tracking is disclosed. The method may include receiving a first image captured by a first camera based at least in part on reflections of light emitted by a light source. The reflections may be partially occluded by an electronic display disposed in front of the first camera. The method may further include detecting an eye of a user in the partially occluded first image and determining a position or orientation of the eye relative to the electronic display.
[030] ... In some embodiments, a neural network model (e.g., a convolutional neural network (CNN)) or other algorithm may be used to filter out noise or interference, such as a “screen door effect,” from the occlusions by the display pixels and/or display subpixels. The resulting image may be used for eye tracking or other applications.
[0120] For example, the processor 1120 may execute the image filtering SW module 1132 to filter images received via the camera interfaces 1114 and/or 1116 . In executing the image filtering SW module 1132 , the processor 1120 may use the neural network model 1133 or other algorithm to filter, reduce, or eliminate noise (such as a screen door effect) or interference from images received via the camera interfaces 1114 and/or 1116 .
Is there anything in the video which suggests a hardware digital SNN?
FrederikSchack
Regular
We could add Synsense, Hailo and Rain. Maybe Qualcomm?
I think NeuronFlow isn't a company, it's GrAI Matter that makes NeuronFlow.
How many of those sell IP blocks (crucial to IoT and AIoT) that is implementable in other chips?
How many of them do Vision Transformers and TENN?
How many have one shot learning?
I think we will not have much direct competition for some years at least.
robsmark
Regular
I think that would depend on how the NDA was written. It could be that once the initiator (let’s assume that is the client) has broken the NDA, then it become obsolete.
I’m keen to see the companies response here.
Makeme 2020
Regular
|
|---|
Reduce development time with the following handpicked reference solutions. The low power consumption performance of the RL78 MCU extends battery life and drives various applications using a small amount of energy. For more information on each reference solution, please click the following links.
| ||
|---|---|---|
Reduce Development Time with RL78 Reference Solutions |
The Director of Global IR would probably be flooded with emails today, not from me though.
I was told that once Mercedes mentioned us at last years CES we could comment, as the NDA had in fact appeared to have been ........
but this comment turned out to be false.
I expect that one line on Teksuns website will be removed, lets see, that way the company aren't required to make any comment
and nor is Teksun.
My view only.
Steve10
Regular
In April, 2018 BRN announced Bager of Southern California as exclusive sale representative for southern California.
brainchip.com
Bager Southern California has been in business since 1980 and has increased its sales every year, including during the major downturns in the semiconductor industry. It has three sales offices, located in Los Angeles, Orange County, and San Diego. BrainChip adds to Bager’s prestigious line card, which already includes premier high technology companies such as Toshiba, Delta Electronics, Epson, and Lattice Semiconductor to name just a few. Bager’s unprecedented longevity equates to deep ties with all of the major OEMs in the region.
Bager might be the link between BRN & Toshiba.
www.bagersales.com
Jason George from Bager liked the Akida Gen 2 article.
BrainChip Inks Bager Southern California as Exclusive Sales Representative for Southern California
BrainChip Inks Bager Southern California as Exclusive Sales Representative for Southern California April 18, 2018
brainchip.com
Bager Southern California has been in business since 1980 and has increased its sales every year, including during the major downturns in the semiconductor industry. It has three sales offices, located in Los Angeles, Orange County, and San Diego. BrainChip adds to Bager’s prestigious line card, which already includes premier high technology companies such as Toshiba, Delta Electronics, Epson, and Lattice Semiconductor to name just a few. Bager’s unprecedented longevity equates to deep ties with all of the major OEMs in the region.
Bager might be the link between BRN & Toshiba.
Bager Sales | Innovative Technology Solutions
Our company represents leading edge technology companies across a wide range of products. Through relationship-driven selling, we have built lasting partnerships based on trust, integrity, and mutual success spanning over 40 years. Our commitment to excellence ensures that our manufacturing partners
www.bagersales.com
About us
Bager of Southern California (BSC) was founded in 1980, to serve the needs of the electronics market place. Whether that need is for Components, Systems, or support in Design and Manufacturing services, BSC provides value driven solutions. Our methodology has resulted in a stable Line Card, loyal employees, long term success, and most importantly, customer satisfaction. BSC covers Southern California including the Mexico border region.Jason George from Bager liked the Akida Gen 2 article.
Jason George
Bager of Southern California
Ventura, California, United States
348 followers 349 connections
About
Experienced Sales Specialist with a demonstrated history of working in the semiconductors industry. Strong sales professional skilled in Semiconductors, Sales, Business Development, Sales Operations, and Integrated Circuits (IC).
Makeme 2020
Regular
Makeme 2020
Regular
BRAINCHIP IS MENTIONED AS ARM PARTNER
Yoda
Regular
Share price aside (which is a temporary malaise) I'd say things are going pretty well wouldn't you? It's increasingly clear that Brainchip's ambition for Akida to be the ubiquitous standard is slowing taking shape. Can there really be any doubt of substantial, I mean really substantial, revenue in the not-so-distant future? We can all see that on the horizon right? I certainly want to be around for that and confess I've even bought some more. 
stuart888
Regular
Happy times for Spiking Smarts, yeah Brainchip!Share price aside (which is a temporary malaise) I'd say things are going pretty well wouldn't you? It's increasingly clear that Brainchip's ambition for Akida to be the ubiquitous standard is slowing taking shape. Can there really be any doubt of substantial, I mean really substantial, revenue in the not-so-distant future? We can all see that on the horizon right? I certainly want to be around for that and confess I've even bought some more.
FabricatedLunacy
Regular
49% down in my portfolio…
Time to not look at the account for a while. I have confidence this will go up and beyond my modest expectations. But days like the last few have been interesting for ones conviction. I’m still 100% sure I make money back however. So it’s just a step back. Breath. And move on.
Time to not look at the account for a while. I have confidence this will go up and beyond my modest expectations. But days like the last few have been interesting for ones conviction. I’m still 100% sure I make money back however. So it’s just a step back. Breath. And move on.
If the three partnerships mentioned by Teksun are true, would not disclosing them now mean many people will remain privy to some pretty hot inside information? I mean finding out that Brainchip has partnered with Toshiba to integrate Akida into its products is reason enough for me top up on BRN. Toshiba sure do sell a lot of appliances that could benefit from Akida’s magic.
GazDix
Regular
Highlights from this podcast for me:
Jeff Herbst realized AI will take off and left Nvidia to go private with a partner.
'This will continue to go up and to the right' (Jeff talking about AI).
'You and I talk about this all the time' (Sean Hehir and Jeff obviously have a close relationship).
I like how they both explain how important the ecosystem is. 'We care about solutions and solving problems and not really the technology' (Herbst). This also helped me better understand why Brainchip were making all these partners.
Hehir specifically mentioned the automotive industry (in the last podcast Nandan mentioned this as well).
I felt like Jeff was talking to also to shareholders when he said you need patience from his experience at Nvidia.
He said:
‘AI is the most fundamentally most disruptive force anyone has seen in our lifetime’.
Paraphrasing this one:
‘Inference side is Brainchip’s competitive advantage.’
Hehir said: ‘We will have some interesting <stuff>, err comments coming out in the near future’ (regarding transformer models).
They both think that 2023 is a big year for the industry led by ChatGPT.
Jeff Herbst realized AI will take off and left Nvidia to go private with a partner.
'This will continue to go up and to the right' (Jeff talking about AI).
'You and I talk about this all the time' (Sean Hehir and Jeff obviously have a close relationship).
I like how they both explain how important the ecosystem is. 'We care about solutions and solving problems and not really the technology' (Herbst). This also helped me better understand why Brainchip were making all these partners.
Hehir specifically mentioned the automotive industry (in the last podcast Nandan mentioned this as well).
I felt like Jeff was talking to also to shareholders when he said you need patience from his experience at Nvidia.
He said:
‘AI is the most fundamentally most disruptive force anyone has seen in our lifetime’.
Paraphrasing this one:
‘Inference side is Brainchip’s competitive advantage.’
Hehir said: ‘We will have some interesting <stuff>, err comments coming out in the near future’ (regarding transformer models).
They both think that 2023 is a big year for the industry led by ChatGPT.
Fact Finder
Top 20
Back home on PC.
Much criticism has been levelled at Sean Hehir for not communicating with shareholders over the last several months and to some extent this type of criticism has been made since shortly after he took the reins at Brainchip.
My intention is not to debate the merits of the criticism suffice to say I on two occasions have made clear a concern I had about his communication style, but as I am just an old technophobe without any experience of running, for the sake of this discussion a Silicon Valley type technology start-up that originated from Australia currently listed on the ASX and intending to eventually launch on Nasdaq during a time of global turbulence with the major shareholders and founders looking over my shoulder, so I expressed those concerns directly to the company.
In expressing those concerns it was emphasised that the company was in the process of reinventing its corporate branding as a commercial entity and while it appeared to be taking backward steps these measures were essential if the end goals were to be realised.
I probably do not need to write that Tony Dawe has been under almost constant attack for the best part of six months from shareholders concerned about the direction of the company and that over this time the company has only poorly at one level communicated the importance of what was taking place to set the foundations for success.
These foundations can be summed up from the company's point of view in the one word "ECOSYSTEMS".
So if we look at the avalanche of news over the last couple of weeks, today's promise by Sean Hehir of more to come, and the focus of today's podcast fronted by Sean Hehir being "ECOSYSTEMS" and the qualifications of the guest to speak precisely on this point it is my impression that this podcast was Sean Hehir trying to explain to shareholders why the pain of the last 18 months since his appointment had been necessary.
I posted earlier a quote relating to competitors but the overall paper is about whether the world is ready for neuromorphic computing and in essence it does nothing if not make clear that in the absence of what Sean Hehir and his guest describe as "Ecosystems" an AKIDA technology revolution cannot take place.
Building these "ECOSYSTEMS" is not sexy, it does not bring in immediate income but without technology "ECOSYSTEMS" AKIDA will have no value to anyone.
Sean Hehir used at the AGM last year his three legged chair to describe what he was doing and how without one of the legs a three legged chair cannot stand or be of any use.
"ECOSYSTEMS" were one of the legs of this chair and today was Sean Hehir attempting to explain with the assistance of someone who has done it for Nvidia why taking the time to build "ECOSYSTEMS" to support AKIDA technology just had to be accomplished otherwise failure would be the outcome.
I hope that this podcast has drawn a line in the sand and that going forward we will see a change in the communication that takes place with shareholders and in this regard I am not talking about lots of IP licence agreements being announced what I am talking about is a less introspective style, one where Sean Hehir takes on fully the role of visionary CEO leading from the front confident that all the foundations needed including "ECOSYSTEMS" are firmly in place.
If this all sounds a bit Tesla-ish having a vision that shareholders, customers and "ECOSYSTEM" partners can embrace is in my opinion absolutely essential if Brainchip is to lead the Artificial Intelligence revolution and have the world follow AKIDA technology science fiction forward into the 4th Industrial Revolution.
Well you did ask.
My opinion only DYOR
FF
AKIDA BALLISTA
Last edited:
chapman89
Founding Member
Sean promised
Super sleuthing Steve,Synaptics is using DSP Group’s nNet Lite NN processor.
DBM10L
DSP Group’s DBM10L is an ultra-low-power, small-form-factor, cost-effective artificial intelligence (AI) and machine learning (ML) SoC based on a digital signal processor (DSP) and neural network (NN) engine, both optimized for voice and sensor processing. It is suitable for battery-operated devices such as smartphones, tablets, wearables, and hearables, including true wireless stereo (TWS) headsets, as well as smart home devices such as remote controls. The DBM10L can enable AI/ML, voice, and sensor fusion functions that include voice trigger (VT), voice authentication (VA), voice command (VC), noise reduction (NR), acoustic echo cancellation (AEC), sound event detection (SED), proximity and gesture detection, sensor data processing, and equalization.
The DBM10L’s NN engine comprises DSP Group’s nNet Lite NN processor, a standalone hardware engine that is designed to accelerate the execution of NN inferences. nNet Lite provides the DBM10L its ML capability and is optimized for maximum efficiency to ensure ultra-low power consumption for small- to medium-size NNs.
The DBM10L is supported by embedded memory, as well as serial and audio interfaces for communication with other devices in the system, such as an application processor (AP), codecs, microphones, and sensors.
DBM10L - DSP Group
DSP Group’s DBM10L is an ultra-low-power, small-form-factor, cost-effective artificial intelligence (AI) and machine learning (ML) SoC based on a digital signal processor (DSP) and neural network (NN) engine, both optimized for voice and sensor processing. It is suitable for battery-operated...www.dspg.com
DSP Group Unveils DBM10 Low-Power Edge AI/ML SoC with Dedicated Neural Network Inference Processor
January 7 2021, 13:00
DSP Group announced the DBM10, a new low-power, cost-effective artificial intelligence (AI) and machine learning (ML) system-on-chip (SoC). This new open platform, with a cost- and power-optimized architecture, enables rapid development of AI and ML applications for mobile, wearables, hearables, and connected devices in general. It provides a complete platform in terms of voice and audio processing, without compromising the battery life of new designs, and allowing developers to implement their own differentiating algorithms.
DSP Group is a global provider of wireless and voice-processing chipset solutions with extensive experience in voice implementation and an increasing focus in advanced audio processing for personal audio with hearables (headphones, headsets, earbuds) and wearables (on-body electronics). Its new DBM10 SoC comprises a digital signal processor (DSP) and the company’s nNetLite neural network (NN) processor, both optimized for low-power voice and sensor processing in battery operated devices.
This dual-core architecture offers developers with full flexibility of partitioning innovative algorithms between DSP and NN processor and enables fast time to market for integration of voice and sensing algorithms such as noise reduction, AEC, wake-word detection, voice activity detection and other ML models.
The DBM10 features an open platform approach with a comprehensive software framework. This allows developers to quickly get next-generation designs to market with their own algorithms, or with DSP Group’s comprehensive and proven suite of optimized algorithms for voice, sound event detection (SED), and sensor fusion, as required by applications ranging from true wireless stereo (TWS) headsets to smartphones, tablets, wearables, and connected devices.
"Edge applications for AI are many and diverse, but almost all require the ultimate in terms of low power, small form factor, cost effectiveness, and fast time-to-market, so we are very excited about what the DBM10 brings to current and new customers and partners," says Ofer Elyakim, CEO of DSP Group. "Our team has worked to make the absolute best use of available processing power and memory for low-power AI and ML at the edge — including developing our own patent-pending weight compression scheme —while also emphasizing ease of deployment. We look forward to seeing how creatively developers apply the DBM10 platform."
The DBM10 adds to DSP Group’s SmartVoice line of SoCs and algorithms that are deployed globally in devices ranging from smartphones and laptops/PCs, to set-top boxes, tablets, remote controls, and smart IoT devices for the home. In 2020, SmartVoice shipments reached the 100 millionth milestone, and the new low-power DBM10 is already supported by an established ecosystem of third-party algorithm providers. Some of these have already begun running their NN algorithms on the nNetLite NN processor at the heart of the DBM10 to achieve maximum performance at the lowest power consumption.
View attachment 32164
Working alongside a programmable low-power DSP, the nNetLite processor supports all standard deep NN (DNN) and ML frameworks and employs a comprehensive cross-platform toolchain for model migration and optimization. The SoC device is supplied in a highly-compact form factor (~4 mm2), specified to support ultra-low-power inference at ~500 μW (typical) for voice NN algorithms and being able to run Hello Edge 30-word detection model @ 1 MHz (125 MHz available) as a reference. The DBM10 allows porting of large models (10s of megabytes) without significant accuracy loss using model optimization and compression.
www.dspg.com
DSP Group Unveils DBM10 Low-Power Edge AI/ML SoC with Dedicated Neural Network Inference Processor
DSP Group announced the DBM10, a new low-power, cost-effective artificial intelligence (AI) and machine learning (ML) system-on-chip (SoC). This new open platform, with a cost- and power-optimized architecture, enables rapid development of AI and ML applications for mobile, wearables, hearables...audioxpress.com
Hisense Selects Synaptics’ DBM10L Processor For First AI-Enabled Always-On Voice Remote Control
AMSTERDAM, The Netherlands, Sept. 09, 2022 – Synaptics® Incorporated (Nasdaq: SYNA) today announced that Hisense, a global leader in consumer electronics and home appliances, selected the DBM10L with its dedicated neural processing unit (NPU) to implement the first artificial intelligence (AI)-enabled always-on voice (AOV) remote control unit (RCU), the EFR3B86H. Hisense paired the DBM10L-equipped RCU with its state-of-the-art 65A9H 4K OLED TV, where Synaptics' high-performance edge-AI processing and low power are vital to ensure the ultimate AOV end-user experience.
“Hisense consistently stays ahead of the curve when it comes to enabling innovative and intuitive features,” said Venkat Kodavati, SVP and Chief Product Officer at Synaptics. “With end users’ increasing reliance upon voice and voice assistants such as Alexa, we are very excited to have worked with them to bring that same experience to TV remote controls. Our collaboration on a high-performance AOV implementation creates the opportunity for remotes to now become a more integral and critical user-engagement platform for the smart home.”
While a reliable and responsive AOV experience for remote controls is increasingly desirable, it is challenging to execute in battery-driven applications. “This is particularly true in noisy environments as more noise translates to more power consumption to prevent performance degradation,” said Shay Kamin Braun, Director of Product Marketing at Synaptics.
The DBM10L enables a superior AOV user experience that combines high performance with ultra-low power consumption, allowing devices to operate for extended periods using a single pair of AAA batteries. “Along with upcoming innovations such as biometrics for voice authentication for online purchases, AOV remote controls for TVs and other consumer devices can now provide greater convenience for users and higher attachment rates for equipment and service providers,” said Kamin Braun.
The DBM10L AOV solution
To solve the power consumption challenge while delivering the best performance for AOV applications, Synaptics built an ultra-low-power voice engine around its DBM10L system-on-chip (SoC), which combines the dedicated NPU with a low-power DSP. The solution comprises the DBM10L and proven algorithms for filtering, noise suppression, beamforming, wake word detection, and voice activity detection. Optimizations allow deep neural network (DNN)-based wake-word detection and other edge AI algorithms to run on the DBM10L's NPU, targeting high performance at ultra-low power with low latency, while different voice and audio processing algorithms run optimally on the integrated low-power DSP.
Availability
The EFR3B86H AOV remote control is shipping now with the Hisense TV model 65A9H. For more information on the DBM10L, visit the DBM10L webpage or contact your local Synaptics sales representative.
For more about the potential of AOV RCUs and how they are changing how we interact with home devices, see “Always-On Voice Makes Content Control Seamless and Intuitive”.
About Hisense
Founded in 1969, Hisense is one of the largest consumer electronics and home appliances companies in the world. Hisense offers a broad range of technology-driven products that are manufactured and distributed across the world, including smart TVs, smart phones, refrigerators, freezers, and air conditioners, among other products. Hisense has a workforce of over 70000 worldwide, and its flat-panel TV market share in China has been No.1 for 13 consecutive years. Currently, Hisense boasts several subsidiaries, with sales revenue reaching CNY 100.3 billion in 2016. For more, visit www.hisenseme.com.
Press Release | Hisense Selects Synaptics’ DBM10L Processor For First AI-Enabled Always-On Voice Remote Control | Synaptics
AMSTERDAM, The Netherlands, Sept.www.synaptics.com
DSP Gp's NN uses single bit synchronous rate coding with leaky-integrate-and-fire;
US11087205B2 Neural cell and a neural network
US201762449129P·2017-01-23; US201815877459A·2018-01-23
[0043] The neuron cell has the following features:
a. The neuron cell is a synchronous digital design that may be operated on a single clock as a time-base.
b. The neuron cell output spike may be synchronized with the system clock and may last a single cycle.
c. The neuron cell “output level” may be defined linearly by its spike rate which ranges between 0 and a global maximum spike-rate (which is an integer division of the system clock frequency), which represents the level 1. During each second the neuron cell performs multiple processing iterations and the outcome of each processing iteration may be a single output pulse. The spike rate is the number of output pulses per second.
d. The inter-connection between two neuron cells within the network may be via a single wire.
e. Synapses weights are multi-bit words and are kept in a neuron dedicated memory (such as a ROM or non-volatile memory)
f. Each neuron cell may incorporate a leaking integrator with a time-constant that is controlled by a parameter alpha (α) that may be predefined and/or may be changed.
g. A pseudo-random generator logic that exists inside each neuron facilitates the implementation of smooth activation function and spreading the spike timings in a minimum cost of silicon area.
h. The neuron activation function defines the average spike rate at the neuron output versus the sum of its internal integral value and the bias theta. One or more different known activation functions (for example between one to three) can be selected, based on the nature of a random variable probability distribution function. Since the activation functions that are implemented are known in the literature, existing Deep-Learning tools can be used for designing the whole network. There is no need to make any modifications to these Deep-Learning tools in order to fit them to the nature of the proposed h/w neuron.
i. A neuron network that includes the neuron cell occupies relatively small silicon area, runs on a slow clock, and consumes ultra-low power.
That's a bit early in the production line to detect me.
Evermont
Stealth Mode
Also from Teksun partner page.
‘MegaChips is a transformational deal for BrainChip, and I’m not sure if the market fully appreciates that yet, but they will in time’
I sense the 'time' referred to is fast approaching.
NB - great post above @Fact Finder. Brakes would appear to be off now.
‘MegaChips is a transformational deal for BrainChip, and I’m not sure if the market fully appreciates that yet, but they will in time’
I sense the 'time' referred to is fast approaching.
NB - great post above @Fact Finder. Brakes would appear to be off now.
Fact Finder
Top 20
Fingers so tightly crossed that if I could get them any tighter they would be out of place but lets hope that is the case.
And as for MegaChips a few of us have been beating that drum to a deaf audience for months and months. Let us hope that they will finally hear.
My opinion only DYOR
FF
AKIDA BALLISTA
Similar threads
