Brianchip??
🎧 aTENNuate: speech enhancement like never before. | Rudy Pei
🎧 aTENNuate: speech enhancement like never before. In real-time, on raw waveforms, and lightning fast ⚡ Speech enhancement is a very difficult task that requires an incredible amount of domain expertise. There are typically many moving parts and expertly crafted pipelines for processing audio...www.linkedin.com
View attachment 68983aTENNuate: Optimized Real-time Speech Enhancement with Deep SSMs on Raw Audio
We present aTENNuate, a simple deep state-space autoencoder configured for efficient online raw speech enhancement in an end-to-end fashion. The network's performance is primarily evaluated on raw speech denoising, with additional assessments on tasks such as super-resolution and...arxiv.org
View attachment 68985
BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
Attachments
Yes. I'm really chuffed they have seen the light on software. As I've said before, BRN were in software years ago with Brainchip Studio. It's the fastest way to market. More maintenance, but ongoing income. There are one or two quite successful software compaies about if you look for them.
DingoBorat
Slim
I don't think a name change to "BrainSoft" is at all advisable though
...Quite right! I was thinking of putting it the other way round.I don't think a name change to "BrainSoft" is at all advisable though
Last edited:
I think I can speak for us all when I say I don't understand what all the fuss is about:
arxiv.org
Raw Speech Enhancement with Deep State Space Modeling
Yan Ru Pei, Ritik Shrivastava, FNU Sidharth
We introduced a lightweight deep state-space autoencoder, aTENNuate, that can perform raw audio denoising, superresolution, and de-quantization. Compared to previous works, the key features of this network are: 1) consisting of statespace layers that can be efficiently trained and configured for inference, 2) allowing for real-time inference with low latency, 3) architecturally simple and light in parameters and MACs, 4) capable of processing raw audio waveforms directly without requiring pre/post-processing, and 5) highly competitive with other speech enhancement solutions.
aTENNuate: Optimized Real-time Speech Enhancement with Deep SSMs on Raw Audio
We present aTENNuate, a simple deep state-space autoencoder configured for efficient online raw speech enhancement in an end-to-end fashion. The network's performance is primarily evaluated on raw speech denoising, with additional assessments on tasks such as super-resolution and...
Raw Speech Enhancement with Deep State Space Modeling
Yan Ru Pei, Ritik Shrivastava, FNU Sidharth
We introduced a lightweight deep state-space autoencoder, aTENNuate, that can perform raw audio denoising, superresolution, and de-quantization. Compared to previous works, the key features of this network are: 1) consisting of statespace layers that can be efficiently trained and configured for inference, 2) allowing for real-time inference with low latency, 3) architecturally simple and light in parameters and MACs, 4) capable of processing raw audio waveforms directly without requiring pre/post-processing, and 5) highly competitive with other speech enhancement solutions.
Last edited:
Brainchip gets a nice mention.
Says Brainchip could be positioned as a global leader in next generation AI hardware.
Worth a read.
Says Brainchip could be positioned as a global leader in next generation AI hardware.
Worth a read.
Space Cadet
Regular
MDhere
Top 20
I agree the denoise capabilityis outstanting. At the agm when they did a demo, i had my hearing aids on and when they played the sentence with the words and noise then they played it with the denoise function i heard every word. Its not if but when. ❤🎧 aTENNuate: speech enhancement like never before. | Rudy Pei
🎧 aTENNuate: speech enhancement like never before. In real-time, on raw waveforms, and lightning fast ⚡ Speech enhancement is a very difficult task that requires an incredible amount of domain expertise. There are typically many moving parts and expertly crafted pipelines for processing audio...
View attachment 68983aTENNuate: Optimized Real-time Speech Enhancement with Deep SSMs on Raw Audio
We present aTENNuate, a simple deep state-space autoencoder configured for efficient online raw speech enhancement in an end-to-end fashion. The network's performance is primarily evaluated on raw speech denoising, with additional assessments on tasks such as super-resolution and...
View attachment 68985
Last edited:
Pom down under
Top 20
Should be a mention for brainchip but not watched it.
Last edited:
Yep, we’re the ‘go to’ outstanding winner, having watched it.
Pom down under
Top 20
Did it mention us?
TheUnfairAdvantage
Regular
Akida 1000 was the star hardware used in their study and it passed with flying colours. They did say in the concluding statements that SNN hardware has some limitations but I imagine the current Akida hardware/software was designed to overcome those limitations. lGreat exposure for Akida. Good work Pom.
I wonder if it can improve a cochlear implant. If it does, surely they are taking note.
Similar threads
