A new weapon has been added to the AI ​​voice market: the online hybrid voice module has been launched

Publisher:EtherealJourneyLatest update time:2019-11-13 Source: 全志科技 Reading articles on mobile phones Scan QR code
Read articles on your mobile phone anytime, anywhere

Recently, SoundAI and Allwinner Technology jointly launched an offline and online hybrid voice module, which helps industry customers upgrade to intelligence with mature and efficient solutions, bringing new surprises to the AI ​​voice module market.

 

          image.png

 

Offline and online hybrid voice module

 

The offline and online hybrid voice module of SoundAI Technology is built on the Allwinner R328 chip. It modularizes SoundAI's industry-leading voice interaction technology and can deliver a complete set of mass-produced intelligent upgrade solutions to industry customers. The module provides a variety of flexible microphone array designs such as 2-microphone, 4-microphone and 8-microphone, integrates AI algorithms such as noise suppression, echo cancellation, and sound source localization, supports far-field voice wake-up, far-field voice recognition, far-field voice synthesis, natural language processing, offline command word wake-up and other capabilities, and is suitable for smart speakers, white appliances, conference machines, cameras, tablets and other devices in multiple scenarios such as smart homes and smart offices.

 

SoundAI's online and offline hybrid voice module adopts a combination of online and offline voice recognition, and has many advantages such as high cost performance, leading voice interaction capabilities, and strong compatibility.

 

In terms of voice algorithms, echo cancellation, noise suppression, sound source localization, reverberation elimination, beamforming and other algorithms are used to solve noise interference in real application environments, while also supporting customization of wake-up words.

 

In terms of voice interaction, it is connected to the full-stack voice channel, covering the entire chain of intelligent interactions such as microphone array, noise reduction wake-up, offline command word recognition, speech recognition, and speech synthesis. Equipped with SoundAI Azero intelligent operating system, it integrates more than 200 common skills and services such as information query, audio and video entertainment, and IOT control based on the full-stack voice interaction technology capabilities. At the same time, it provides simple and easy-to-use voice skill development tools and one-stop intelligent voice software and hardware solutions, which can greatly reduce the development difficulty and usage threshold of AI industry applications, and quickly meet the personalized needs of different users and industry groups. In terms of industry applications, six versions have been launched to meet the different needs of ecological customers, industry customers, operators, and white goods customers, including the front-end SDK version, system version, operator customized version, Amazon AVS customized version, white goods online hybrid version, and domestic ecological customized version. You can flexibly choose the corresponding solution according to actual needs.

 

image.png

 

SoundPi Mini Board

 

Based on the offline and online hybrid voice module, SoundAI Technology simultaneously launched SoundPi Mini Board, a hardware and software integrated development board. SoundPi Mini Board provides developers and small and medium-sized enterprise customers with a low-cost voice entry threshold, integrating the offline and online hybrid voice module and the Azero intelligent operating system, supporting barrier-free voice interaction within a range of 5 meters, and providing rich service support. Based on this, the development and application of intelligent voice products can be carried out simply and quickly.

 

In addition, the R328 chip used in the module is a dedicated processor for intelligent voice developed by Allwinner Technology. The chip adopts the industry's extremely mature and reliable A7 1.2G dual-core design, built-in high-performance 3-way ADC, 8-way DMIC, built-in 64MB/128MB DDR different capacity versions, built-in EQ adjustment, sound effect algorithm, rich digital audio input and output interfaces and flexible microphone array interface. At the same time, the chip adopts 28nm process technology, mad-vad, small chip size, low power consumption and low heat generation characteristics.

 

The strong combination of SoundAI's "soft power" and Allwinner's "hard power" has created the high performance and low power consumption of the offline and online hybrid voice module. In the future, the two parties will continue to explore deeper cooperation in more fields and launch more high-quality products and services for global users.

 

About SoundAI Technologies

 

SoundAI Technology is a platform service provider that integrates acoustic and artificial intelligence technologies. It is committed to realizing the mission of "making machines smarter" and achieving the vision of "improving life with technology". It mainly provides SoundAI Azero intelligent operating system and services, as well as artificial intelligence technologies and product solutions that are deeply integrated with application scenarios. It is typically used in smart homes, smart conferences, smart education, smart medical care, smart manufacturing, smart cars, financial technology, energy and electricity, smart courts, smart transportation and robots. It has independent core technologies such as acoustic vector sensing, acoustic array chips, acoustic structure design, far-field acoustic processing (echo cancellation, noise suppression, sound source localization, reverberation elimination, beamforming, etc.), far-field voice wake-up, far-field voiceprint recognition, far-field speech recognition, far-field duplex call, natural language understanding, natural speech synthesis, ultra-far-field acoustic monitoring, ultra-near-field voice interaction, distributed field voice interaction, deep learning and small sample learning, intelligent search and personalized recommendation, knowledge graph and intelligent analysis, and accurate user spatiotemporal portrait.

 

About Allwinner Technology

 

Allwinner Technology was founded in 2007 and is a leading design company for intelligent application processor SoC, high-performance analog devices and wireless interconnect chips. Headquartered in Zhuhai, China, it has R&D centers or branches in Shenzhen, Hong Kong, Xi'an, Beijing, Shanghai and other places. Allwinner Technology is customer-centric, gathers an excellent team, insists on long-term investment in core technologies, and provides market-leading system solutions and considerate services in ultra-high-definition video encoding and decoding, high-performance CPU/GPU/AI multi-core integration, high integration of advanced processes, ultra-low power consumption, and full-stack integrated platforms. The products are widely used in smart hardware, tablets, smart home appliances, Internet of Vehicles, robots, virtual reality, network set-top boxes, power analog devices, wireless communication modules, smart Internet of Things and other product fields.

Reference address:A new weapon has been added to the AI ​​voice market: the online hybrid voice module has been launched

Previous article:Hetai 2019 Product Tour—Looking to the Future, Smart Life
Next article:Quan Ding launches dual-mode fully automatic smart door lock solution based on Qualcomm QCA4024

Latest Internet of Things Articles
Change More Related Popular Components

EEWorld
subscription
account

EEWorld
service
account

Automotive
development
circle

About Us Customer Service Contact Information Datasheet Sitemap LatestNews


Room 1530, 15th Floor, Building B, No.18 Zhongguancun Street, Haidian District, Beijing, Postal Code: 100190 China Telephone: 008610 8235 0740

Copyright © 2005-2024 EEWORLD.com.cn, Inc. All rights reserved 京ICP证060456号 京ICP备10001474号-1 电信业务审批[2006]字第258号函 京公网安备 11010802033920号