Mass production quality of TinyML through C code generation of trained models and memory layout optimization.
Converts Keras/TFLite/ONNX to C code for STM32 using STM32Cube.AI. It optimizes weight placement for SRAM, Flash, and external QSPI, and explains MPU/D-Cache settings and STM32N6 NPU based on primary sources AN5050/UM2526. For more details, please refer to the technical column in the related links.
Inquire About This Product
basic information
【Technical Elements】 - STM32Cube.AI - Keras / TFLite / ONNX conversion - SRAM / Flash / QSPI weight placement optimization - MPU / D-Cache configuration - STM32N6 NPU / AN5050 / UM2526 【Category】Embedded / TinyML / Edge AI *For more details, please refer to the technical column in the related links.
Price range
Delivery Time
Applications/Examples of results
【Intended Use】 - Mass production implementation of AI inference (TinyML) on microcontrollers - Development of edge AI devices - Embedding of pre-trained models 【Examples of Achievements】 - Model C code generation and memory layout optimization using STM32Cube.AI
Company information
Technosphere Co., Ltd. is a system development company based in Osaka that tackles customer challenges in advanced technology areas such as AI, IoT, and web system development. Since its founding in 2021, the company has leveraged a flat team structure to achieve a flexible and speedy development style. We are engaged in solutions that directly address social issues, such as image inspection AI and smart factory support systems.
