BRN Discussion Ongoing

[Pom down under]

Just all this business with Shaw Brothers, shorting and the Shaw Brothers connection with our new financiers et all.

I did post yesterdey that the 50 mill that disappeared from the shorts overnight could have been used in this way and now it’s making a bit more sense. Be good to find out if it was the case

 

[Pom down under]

Mistakes happen, but, it's not a good look.
Particularly in sales, the adage is, that you only get one chance at a first impression.
These stupid errors keep on occurring and they should be getting this stuff proofed, before it's released.
Particularly when we are prowling for the eye's of a partner who would then be contemplating getting into bed with us, and will be relying upon us to provide efficient and effective implementation and ongoing support.

I could do a better job and that’s really saying something

 

[Tothemoon24]

FORBESINNOVATIONAI

Real-Time AI At The Edge May Require A New Network Solution

using TENNs for object detection can provide a 5x to 30x reduction in operations using 1/50th the number of parameters with equal or better accuracy than traditional CNNs

www.forbes.com

Real-Time AI At The Edge May Require A New Network Solution​

Jim McGregor
Contributor
Tirias ResearchContributor Group
Follow





0
Aug 2, 2024,11:10am EDT

The complexity of edge networks
BRAINCHIP
AI has the power to change every electronic platform, but what works in the data center may not work in an industrial edge platform like a security camera, robotic arm, or even a vehicle. There is no one-size-fits-all for edge AI because of space, power, data security, and performance-latency requirements, which means there is not one solution for all AI applications. Transitioning to edge AI requires new solutions, especially for on-device training.

The Current AI Model​

Most AI models start in the data center with the training of a neural network model based on public and/or private data. As we have argued in the past, the expense of running AI in the cloud can be prohibitive from a cost, latency, and security standpoint. As a result, neural network models can be optimized by shrinking the model size to then run at the edge of the network on the device, commonly referred to as “edge computing.” This optimization is a delicate balance of reducing the size of the model while maintaining acceptable accuracy.


While neural network models and optimization techniques improve, using a scaled-down data center solution may not be optimal for many edge applications. Edge applications are often focused on the input of sensor data and require even smaller models with a high degree of accuracy and real-time, or close to real-time, execution for mission critical or even life-threatening situations. These edge applications may include healthcare, automotive/transportation, manufacturing, and security.

Event-driven AI​

As an alternative, BrainChip has developed its Akida neuromorphic IP (intellectual property) solutions to support Temporal Event-based Neural Networks (TENNs), an event-based neural processing network architecture, in addition to traditional Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNNs) and Transformer based neural networks. What this means is that a temporal (time) enabled neural network (TENN), or networks, is only operating during the time when a trigger event or input occurs. During other times, it is not computing and therefore not consuming much power. This can translate to higher performance, adaptability and lower real-time latency per event, input, or request and at a fraction of the power consumption of other AI solutions.


According to BrainChip, TENNs are ideal for processing various types of data such as one-dimensional time series and spatial-temporal data. TENNs is showing positive results in a number common applications, such as audio denoising, eye-tracking for AR/VR, health data monitoring (heart rate, SpO2), keyword spotting, Small Language Models (SLMs), and video object detection. TENNs ability to adapt to new input/events overcome some of the limitations of traditional neural networks while supporting future neural network models at a fraction of the die area and operating cost.

The Akida technology is available as IP cores that can be integrated into anything from a low-power microcontroller to a high-performance applications processor or system-on-chip (SoC) or discrete accelerator chips. While using an IP solution would require a new design, it provides more efficient processing of TENNs models by taking advantage of sparsity, a key feature that makes the human brain so efficient. Additionally, there are no other off-the-shelf solutions that can provide comparable low-power, high-accuracy, and real-time AI processing.

Edge-Specific AI​

While TOPS (Trillions of Operations Per Second) seem to be getting a lot of attention as a means of measuring the AI performance of a chip, analyzing the performance of Edge AI is better accomplished using metrics that are more specific to the application. Examples of these more application specific metrics include measuring the efficiency of processing frames or frames per second (FPS) for image processing, time to first token (TTFT) for response time, mean average precision (mAP) for accuracy, and Watts or Kilowatts for power consumption. BrainChip has provided some data along these lines to demonstrate the value of TENNs in various applications.

In another demonstration, the company demonstrated how TENNs can be used to drastically reduce the training time by and the power consumed by more than orders of magnitude relative to other large language data sets like GPT-2, which would be more appropriate for embedded applications than the newer models, with equivalent accuracy.

Edge Specific Solutions​

While there is a rush to push AI to every platform and device, scaling down from the data center may not be the best solution for many applications. As we have seen in the past, the unique requirements of devices often drive innovation in new directions, Tirias Research believes the same will hold true for AI as it moves from the datacenter to the edge. But, as with any new technology, success often depends on the benefit over existing solutions. According to BrainChip, the numbers can be very significant, with demonstrations showing up to a 50x reduction in the number of model parameters, up to a 30x reduction in training time, and 5000x reduction in multiple-accumulate (MAC) operations with the same or better accuracy. Improvements in performance and power efficiency scale with model efficiency.

Edge Specific Solutions​

While there is a rush to push AI to every platform and device, scaling down from the data center may not be the best solution for many applications. As we have seen in the past, the unique requirements of devices often drive innovation in new directions, Tirias Research believes the same will hold true for AI as it moves from the datacenter to the edge. But, as with any new technology, success often depends on the benefit over existing solutions. According to BrainChip, the numbers can be very significant, with demonstrations showing up to a 50x reduction in the number of model parameters, up to a 30x reduction in training time, and 5000x reduction in multiple-accumulate (MAC) operations with the same or better accuracy. Improvements in performance and power efficiency scale with model efficiency.

 

[Pom down under]

Episode 34: The State of Neuromorphic Computing

Episode 34: The State of Neuromorphic Computing In this episode of "This is Our Mission" podcast, Sean Hehir and guest, Dr. Eric Gallo, Technology R&D Senior Principal at Accenture Labs discuss the benefits of neuromorphic technology and the current state of edge computing plus advancements...

brainchip.com

 

[CHIPS]

Exactly. Thank you!

 

[CHIPS]

No.

But why not?
I would return your teddy to you almost brandneu looking.

 

[IloveLamp]

I had the great privilege of talking with Sean Hehir about the wide-ranging applications of neuromorphic technology, from wearables to space, for the "This is Our Mission" podcast from BrainChip . | Eric Gallo

I had the great privilege of talking with Sean Hehir about the wide-ranging applications of neuromorphic technology, from wearables to space, for the "This is Our Mission" podcast from BrainChip .

www.linkedin.com

 

[IloveLamp]

Real-Time AI At The Edge May Require A New Network Solution

using TENNs for object detection can provide a 5x to 30x reduction in operations using 1/50th the number of parameters with equal or better accuracy than traditional CNNs

www.forbes.com

 

[Rach2512]

The fact that you judge my knowledge based on how many comments I wrote in here says it all about you Ive been invested since 2020 and follows everything this company is doing. My patience with Sean is gone. He almost had 3 years to make us succeed. The technology should literally sell itself. I respect if people want him to stay but I think new blood is necessary

On ignore you go, bye bye

 

[Smoothsailing]

Would anyone kindly like to discuss if these new chiplet designs could include BRN ?

 

[Pom down under]

The fact that you judge my knowledge based on how many comments I wrote in here says it all about you Ive been invested since 2020 and follows everything this company is doing. My patience with Sean is gone. He almost had 3 years to make us succeed. The technology should literally sell itself. I respect if people want him to stay but I think new blood is necessary

On ignore you go, bye bye

i must have missed this one thanks @Rach2512

 

[Pom down under]

Would anyone kindly like to discuss if these new chiplet designs could include BRN ?

Why don’t you ask the company directly

 

[rgupta]

I did post yesterdey that the 50 mill that disappeared from the shorts overnight could have been used in this way and now it’s making a bit more sense. Be good to find out if it was the case

View attachment 67507

That means shorters buy those shares sold by company. It is good and bad in a sense. Good in the way short is almost over. Bad in the mean they can come back again as they have won the 1st lottery and secondly, management had given a middle finger to long term holders and supported the shorters.
But definitely now the company is becoming professional and start making the same tricks other companies make.
By the way what are the chances management is involved through their own friends in shorting and bringing the sp down and now providing them new shares at almost one third of cost.
Dyor

 

[Smoothsailing]

Why don’t you ask the company directly

View attachment 67516

I did via LinkedIn with no response

 

[Pom down under]

I did via LinkedIn with no response

I know hence the earlier response lol

 

[Pom down under]

That means shorters buy those shares sold by company. It is good and bad in a sense. Good in the way short is almost over. Bad in the mean they can come back again as they have won the 1st lottery and secondly, management had given a middle finger

Does feel that way which is probably not the case which says a lot about management atm

By the way what are the chances management is involved through their own friends in shorting and bringing the sp down and now providing them new shares at almost one third of cost.
Dyor

 

[7für7]

The question is what some small investors are thinking when they puff themselves up in an anonymous forum and demand the resignation of a CEO of a company where they are simply hoping to get rich through the efforts of others (employees) or, depending on how much they have invested, to make some profit. Don't they realize they are overestimating themselves? That it is just ridiculous? There are investors who have invested multiples of what they have and are more relaxed... probably because they haven't put everything on one horse. I don't know how many times I've mentioned it... but because of such annoying small investors, publicly traded companies often like to use tactics to get rid of them through reverse splits... the sad part is that the patient ones get affected too... just hold still or sell if something doesn't suit you... everything else is unnecessary and annoying. If someone think he have the knowledge to change something, feel free to apply as an CEO at brainchip… otherwise STFU

 

[7für7]

That means shorters buy those shares sold by company. It is good and bad in a sense. Good in the way short is almost over. Bad in the mean they can come back again as they have won the 1st lottery and secondly, management had given a middle finger to long term holders and supported the shorters.
But definitely now the company is becoming professional and start making the same tricks other companies make.
By the way what are the chances management is involved through their own friends in shorting and bringing the sp down and now providing them new shares at almost one third of cost.
Dyor

If I remember well, if you take the offer and buy the shares, you have to hold minimum 1 year. So, one year of possible Healy grow with less shorts. ?

 

[DK6161]

Can't wait to see if the SPP offered to us small retail SH will be exhausted. It would be a great sign if it does.
Would also be good to know Sean's movements to get an idea who is he engaging with.

 

[Tothemoon24]

TONOMOUS VEHICLESROBOTICS3D PRINTINGIOTSEMICONDUCTORSAEROSPACEMORE >

The Audio Revolution at the Edge​

Hiba Akbar
01 Aug, 2024
FOLLOW
Sponsored by

BrainChip Audio Denoising solution uses Temporal Event Neural Networks (TENNs) to support better audio quality by removing unwanted voices.​

Artificial Intelligence
- Audio
- Energy Optimization
- Low Power Consumption
This article was first published on
brainchip.com
As our world increasingly relies on sound, the demand for pristine audio quality rises. Clear and crisp audio has become essential for various applications such as voice assistants and video conferencing. However, background noise can often interfere with audio quality, making communication and entertainment less enjoyable. Effective audio denoising is essential for these applications.
BrainChip has developed an Audio Denoising solution that uses Temporal Event Neural Networks (TENNs), a set of algorithms designed to enhance performance and power efficiency.
The Audio Denoising solution stands out by minimizing the need for extensive data conversion that consumes significant power. This efficiency supports better audio quality and aligns with the growing demand for sustainable technology solutions in the audio industry.

The Challenge of Audio Denoising​

Audio denoising has long been a complex challenge in signal processing. Traditional methods often lack balancing noise reduction while preserving sound quality.

• One of the main challenges in audio denoising is distinguishing between desired audio and unwanted noise. This can be especially difficult when the noise has similar characteristics to the audio signal, such as in background conversations or environmental sounds.
• Another challenge is in the computational complexity of audio denoising algorithms. Many traditional methods require significant processing power and memory resources, making them unsuitable for deployment on resource-constrained devices.

These limitations have become problematic as the demand for real-time audio processing grows in various applications.

BrainChip Audio Denoising: A Solution to Unwanted Noise​

BrainChip's Audio Denoising solution enhances audio quality by removing unwanted noise. Powered by TENNs, this solution offers improvements in performance and energy efficiency. This makes it a good option for edge-computing audio applications.

Key Features of BrainChip Audio Denoising​

• The solution includes medium-sized models that achieve an impressive Perceptual Evaluation of Speech Quality (PESQ) score of 3.36 with just 590,000 parameters. This makes it highly scalable to meet specific customer needs.
• The TENN models require fewer computational resources compared to traditional models. They use 12 times fewer multiply-accumulate operations (MACs) and nearly three times fewer parameters. This leads to lower power consumption.
• The Audio Denoising solution comes with integrated Hardware IP. It can easily be incorporated into System-on-Chip (SoC) designs for optimal performance in audio processing tasks.
• It is suitable for a range of devices, including wireless earbuds, VoIP systems, and smart home products. It can also be used as a standalone feature or as part of a more extensive system.
• Customers with a TENNs license can fine-tune the TENN models to cater to specific audio environments. This ensures tailored performance for various use cases.

Benefits of Audio Denoising Solution​

BrainChip Audio Denoising solution provides several important benefits for various applications:

• Improved Sound Quality: This solution enhances audio clarity by removing unwanted noise. This makes conversations, music, and other sounds more enjoyable.
• Better Communication: Denoising improves speech intelligibility in noisy environments, such as busy offices or public places, making conversations easier to understand.
• Enhanced User Experience: Clear audio is important for devices like hearing aids or smart speakers. BrainChip Denoising ensures users can hear content without distractions from background noise.
• Energy Efficiency: The Denoising solution consumes less power. This extends the battery life of devices, which is essential for portable electronics.

Real-World Applications of BrainChip Denoising Solution​

BrainChip Audio Denoising solution can enhance audio quality and improve user experiences in multiple settings.

1. Smart Home Devices​

The solution plays a vital role in smart home technology. By improving the accuracy of voice-controlled systems, it ensures that commands are recognized correctly. This leads to better functionality and user satisfaction.

2. Industrial Monitoring​

In industrial settings, the solution enhances the quality of audio data. This improvement leads to more accurate analysis and predictive maintenance. This offers increased efficiency and safety in operations.

3. Medical Devices​

The technology also applies to medical devices, which aid in vital sign estimation. This capability offers accurate and energy-efficient monitoring of human health, which is important in healthcare settings.

4. Speech Clarity​

The Denoising solution can also be used in voice calls and audio recordings. TENN models remove background noise and allow the speaker's voice to come through clearly. This feature is beneficial for mobile devices and hearing aids, where clear communication is essential.

5. VoIP Communication​

The BrainChip Denoising solution improves audio quality in video conferencing and online meetings. Providing more precise and noise-free audio enhances the experience of remote work. This way, the conversations can be more productive and enjoyable.

Implementation and Future Directions​

BrainChip Audio Denoising solution is also integrated into the Akida 2.0 IP, optimized for TENNs technology. The architecture includes a network of nodes, each equipped with an event-based TENN processing unit. This provides a powerful platform for audio processing tasks.

• Future developments will aim to improve efficiency by enhancing activation sparsity, which will reduce the number of active computations needed.
• There will also be a focus on exploring more flexible model architectures to adapt to different audio processing needs.
• Additionally, integrated solutions will be developed to combine audio denoising with speech recognition and keyword spotting tasks. This will offer a more comprehensive audio processing toolkit.

Final Words​

BrainChip's Audio Denoising Solution is a big step forward in audio processing technology. By managing power consumption and computational demands, the solution supports the development of edge devices that provide clear audio in various environments.
This smart technology can make a big difference in how you enjoy your music, videos, and calls. With less noise and better sound, you can focus on what matters.