Dual-core high-performance main control 240MHZ
Open source AILINKER backend service framework
Dual-channel digital silicon microphone
Support dual-microphone noise reduction algorithm
Support VAD algorithm Support AEC algorithm
Support Access to CHATGPT
Free TECHNICAL SUPPORT AND OPEN SOURCE!
AI_VOICE Z01 is a cost-effective AI audio development kit that can be used to quickly implement A1 voice applications and embedded A1 learning, and quickly develop and verify A1 voice products.
Built-in
l dual-channel high-sensitivity and high-SNR digital microphones,
l DVP universal camera interface,
l TF card slot,
l RGB colorful lights
l 3.7V lithium battery interface
l Battery voltage measurement IC and other peripherals
The ESP32-S3R8/ESP32-S3R2 chip it uses integrates low-power WiFi and BLE5.0, external 16MB FIash and 8MB/2MB PSRAM, built-in hardware encryption accelerator, random number generator, HMAC and digital signature module, meeting users' development needs in low-cost, low-power, A1 large model toys, Internet of Things, mobile devices, outdoor monitoring, smart home and other fields.
l Dual-core high-performance main control 240MHZ
l Open source AILINKER backend service framework
l Dual-channel digital silicon microphone
l Support dual-microphone noise reduction algorithm
l Support VAD algorithm Support AEC algorithm
l Support custom access large model
l CHATGPT, Zhipu Qingyan, Tongyi Qianwen, Wenxin Yiyan, Doubao
Product Features
l Technical support and open source
l Lightweight
l Safe
l Low latency
l Wake-up word customization
l Customized control instructions
l Fast calculation speed
l Local model
l Offline operation
l Quick response
l No network transmission delay
l Personalized customization
l Corpus collection
l Easy to add and delete commands
l Flexible binding actions
l Local voice wake-up
l Low resource consumption of wake-up model
l Front-end acoustic algorithm
l High recognition accuracy
l Can support 5 wake-up words
l Internal RAM: 15~24 KB.
l CPU load 9%~30%(ESP32)
l Microphone array algorithm, echo cancellation,
l Noise reduction, sound activity detection
Amazon certified Audio Front End solution
The AFE algorithm has passed the Audio Front End certification of Amazon Alexa built-in devices. The algorithm is optimized through the AI accelerator of ESP32-S3, and when combined with the hardware, 360-degree sound pickup can be achieved using only two microphones.
