TL;DR:
- Meeami Technologies introduces AI-based noise suppression solutions for the Cadence Tensilica HiFi DSP family.
- Proprietary edge-based deep learning networks identify speech in noisy conditions, delivering clean and distortion-free audio.
- Meeami’s tailored AI solutions optimize performance across various HiFi DSP models.
- The Tensilica HiFi Mini achieves remarkable efficiency with 150 MCPS and a compact 250 KB model size.
- Collaboration with Cadence underscores the efficiency and performance of Tensilica HiFi DSPs.
- Meeami’s solutions target multiple applications, including smart home appliances, TWS, earbuds, headsets, and hearing aids.
- Innovative design techniques and quantization methods reduce AI model size with minimal performance loss.
Main AI News:
In a groundbreaking move within the realm of Audio AI, Noise Cancellation, Speaker ID, and Spatial Audio, Meeami Technologies (Meeami®) has unveiled its revolutionary AI-based background noise suppression solutions. These cutting-edge solutions, specifically designed for the Cadence Tensilica HiFi DSP family, promise to redefine audio quality in the industry.
Meeami’s approach leverages proprietary edge-based deep learning networks to identify and distinguish speech even in the most challenging, noisy conditions. By effectively suppressing background noise and enhancing the clarity of speech, Meeami ensures a distortion-free and immersive audio experience. Notably, Meeami’s speech enhancement and speaker ID solutions have already made their mark, powering over 40 million devices globally.
Meeami’s commitment to innovation is exemplified through its successful optimization of AI footprints, aligning them seamlessly with Cadence’s Tensilica HiFi DSPs. The result is a suite of embedded AI solutions, each meticulously tailored to harness the strengths of individual HiFi DSP models, thus maximizing performance across the entire HiFi DSP family. This approach underscores the unique advantages offered by each DSP model, solidifying the HiFi DSP family’s reputation for versatility and unparalleled performance.
In the case of the Tensilica HiFi Mini, Meeami’s technology sets a new standard for efficiency. Operating at approximately 150 million cycles per second (MCPS) and occupying a mere 250 KB in model size, this achievement highlights Meeami’s unwavering dedication to delivering high efficiency within compact form factors, firmly establishing itself as an industry leader.
Meeami’s collaboration with Cadence was driven by the unmatched performance and efficiency of Cadence’s Tensilica HiFi DSPs. These DSPs offer specialized capabilities such as optimized audio processing, reduced power consumption, and the ability to handle complex audio AI algorithms. Consequently, they emerge as the preferred choice for Meeami’s advanced noise suppression technology. Cadence’s Tensilica HiFi DSPs have already gained widespread adoption among OEMs worldwide, with billions of HiFi DSPs integrated into a diverse range of products.
Yipeng Liu, Product Marketing Group Director for Tensilica audio/voice DSPs at Cadence, highlighted the collaborative innovation, stating, “Partners like Meeami drive innovation in the AI space and highlight the efficiency that our Tensilica HiFi DSPs bring to edge-based and on-device AI applications. We are excited to work with Meeami to bring ultra-low-power noise suppression to enhance user experiences in next-generation audio device communication.”
Harish Rajamani, Director of Engineering at Meeami Technologies, shared his enthusiasm, saying, “We are excited to announce a low-footprint cutting-edge AI solution on Cadence Tensilica HiFi DSPs to improve the user experience for multiple use cases such as voice-controlled smart home appliances, TWS, earbuds and earphones, headsets, and hearing aids.”
Sasank Kottapalli, AI Lead for Speech Products at Meeami Technologies, emphasized the achievement of a substantial reduction in AI model size, coupled with minimal performance loss, through innovative design techniques and quantization methods in deep learning AI models. This achievement reaffirms Meeami’s commitment to advancing the boundaries of audio AI technology while ensuring efficiency and performance remain at the forefront of their solutions.
Conclusion:
Meeami’s AI-driven noise suppression solutions mark a significant leap in audio quality enhancement. Their tailored approach to DSP models and commitment to efficiency position them as industry leaders. This innovation will likely lead to improved user experiences across various audio device markets, reinforcing the demand for advanced noise suppression technology.
