All items tagged with inference (16)

New Expressive Real-Time Text-to-Speech (TTS) Models Support High-Fidelity Voice Cloning

February 17, 2025 by Joao Martins

New Expressive Real-Time Text-to-Speech (TTS) Models Support High-Fidelity Voice Cloning

Zyphra, a company based in Palo Alto, California, building a mew multimodal artificial intelligence agent system, announced the release of Zonos-v0.1 beta, a pair of extremely expressive text-to-speech (TTS) models with high fidelity voice cloning. The company is releasing both the transformer and hybrid TTS models with an Apache 2.0 license, which effectively makes these available for redistribution.

PIMIC and ZillTek Launch Integrated Edge AI and Microphone Noise Cancellation Solution

December 23, 2024 by Joao Martins

PIMIC and ZillTek Launch Integrated Edge AI and Microphone Noise Cancellation Solution

PIMIC, a new company working on ultra-low-power AI technology, unveiled the Clarity NC100, a Deep Neural Network (DNN)-based Environmental Noise Cancellation (ENC) chip that sets a new benchmark for performance and edge AI compute efficiency. Working in partnership with Taiwan-based MEMS microphone developer ZillTek, at CES 2025 the Californian company will present an integrated edge AI chip and digital microphone designed to address noise cancellation challenges in a single package.

Ceva and Edge Impulse Join Forces to Enable Faster Development of Edge AI Applications

September 24, 2024 by audioXpress Staff

Ceva and Edge Impulse Join Forces to Enable Faster Development of Edge AI Applications

Ceva and Edge Impulse announced a collaboration to rapidly develop, train and deploy advanced embedded machine learning (ML) applications on Ceva's neural processing units with cycle-accurate performance, even before physical hardware is available. This first-of-its-kind collaboration based around Ceva's NPU IP ensures that developers can accurately benchmark and refine the performance of their edge AI models using Edge Impulse's platform, significantly accelerating the time-to-market.

December 6, 2023 by audioXpress Staff

Ceva Launches New Brand Identity Reflecting its Focus on Smart Edge IP Innovation

Ceva, Inc., introduced a fresh new brand visual identity, logo design and web domain ”ceva-ip.com,” reflecting the company’s commitment to "being the partner of choice for transformative IP solutions that power the smart edge." Ceva is a leading provider of silicon and software IP that makes it possible for low power Edge AI devices to connect, sense and infer data, reliably and efficiently, across multiple high-growth end markets.

Dolphin Design Unveils Innovative IP for Sound Classification Reducing Power Requirements by 99%

June 27, 2023 by audioXpress Staff

Dolphin Design Unveils Innovative IP for Sound Classification Reducing Power Requirements by 99%

Dolphin Design, a semiconductor designer of power management, audio, and processing devices with in-house ASIC design services, announced the launch of WhisperExtractor, a mixed-signal technology for voice and audio-enabled SoCs, enabling voice and sound classification at a record µW level power budget. The new technology paves the way to breakthrough always-on voice user interface, real time speech-to-text, and always-on sound detection.

Syntiant Launches NDP115 Neural Decision Processor for Always-On Speech and Sensor-Fusion

February 14, 2023 by Joao Martins

Syntiant Launches NDP115 Neural Decision Processor for Always-On Speech and Sensor-Fusion

It was one of the key audio-related demonstrations at the Syntiant suite during CES 2023, and the new NDP115 Neural Decision Processor is now positioned as a key strategic platform for new compact, ultra-low-power designs with always-on voice and sensor fusion applications. The newest addition to Syntiant's family of special purpose silicon is built using the company’s Syntiant Core 2 inference engine that can run multiple neural network loads at under 1mW.

STMicroelectronics Updates NanoEdge AI Studio to Support On-Device Learning and Inferencing in Recently Announced AI-Native MEMS Sensors

June 24, 2022 by audioXpress Staff

STMicroelectronics Updates NanoEdge AI Studio to Support On-Device Learning and Inferencing in Recently Announced AI-Native MEMS Sensors

STMicroelectronics has expanded its NanoEdge AI Studio machine-learning design software with support for smart sensors that contain ST’s embedded Intelligent Sensor Processing Unit (ISPU). The new release extends the market-unique tool’s capability to enable on-device learning of AI models for anomaly detection inside intelligent sensors.

December 7, 2021 by audioXpress Staff

RISC-V International Ratifies 15 New Specifications Opening Up New Possibilities for RISC-V Designs

RISC-V is bringing together different companies and industries, combining a modular technical approach with an open license business model for the RISC-V Instruction Set Architecture (ISA). RISC-V International, the global non-profit open hardware standards organization, incorporated in Switzerland, just announced a new set of specifications that will be key for advanced audio applications using machine learning (ML).

Syntiant Introduces NDP102 Neural Decision Processor for Edge AI Sensor Applications

September 16, 2021 by audioXpress Staff

Syntiant Introduces NDP102 Neural Decision Processor for Edge AI Sensor Applications

Syntiant, a provider of deep learning solutions making edge AI a reality for always-on applications in battery-powered devices, announced its new Syntiant NDP102 Neural Decision Processor (NDP) for sensor applications that delivers 100x improvement in efficiency and performance compared to current MCU-based offerings. The fourth edge AI chip from the Californian company enables highly accurate inference performance at industry leading low power consumption.

Sondrel Unveils Edge Computing IP Platform for Smart Devices

September 8, 2021 by audioXpress Staff

Sondrel Unveils Edge Computing IP Platform for Smart Devices

Sondrel has created a powerful, quad core IP platform that is ideal for ASIC solutions for remote gathering and processing of video and data at the Edge with secure onwards transmission of the results. The SFA 200 reference design is targeted at battery powered applications in Smart Home, Sensor Fusion, voice-control, and other devices where a compact chip can add processing ‘smarts’ by providing local end-point data processing for data collection and analysis along with inference processing.

«
1
2
»