Chipintelli Introduction¶
Company Profile¶
Chipintelli, established in 2015 and headquartered in Chengdu, is a leading AI voice technology company. We specialize in the design and development of AI voice chips and advanced speech processing algorithms. As a full-stack technology innovator, we offer end-to-end solutions encompassing AI chip design, speech processing algorithms, voice application development platforms, and integrated hardware-software systems. Our leadership in the AI voice chip sector is demonstrated by our industry-leading market share in edge-side voice computing solutions.
AI Voice Chips¶
Since its establishment, Chipintelli has launched multiple chips, all featuring high performance, high integration, high reliability, low power consumption, and low cost.
In 2024, Chipintelli introduced two major AI speech chip series: the CI13LC series (including CI13161, CI13162, CI13241, CI13242, CI13322, and CI13162P) and the CI23LC series (including CI23161, CI23162, CI23242). Both series are equipped with the new generation self-developed brain neural network processor BNPU V3.5, with a system frequency of up to 210MHz, built-in 288KByte SRAM, integrated PMU power management unit, RC oscillator, single-channel high-performance low-power Audio Codec, and multiple peripheral control interfaces such as UART, IIC, PWM, GPIO, fully supporting offline AI speech recognition. The CI13LC series’ CI13161 and CI13162 chips are fully compatible with the V3 chips CI1311 and CI1312, requiring no changes to peripheral hardware; CI13162P further integrates a high-performance amplifier unit, enhancing audio playback capabilities. The CI23LC series integrates high-performance low-power Bluetooth BLE, requiring only a few peripheral components to achieve high-performance speech recognition, speech noise reduction, and strong environmental noise suppression capabilities. The CI23162 and CI23242 also support Bluetooth BLE5.4 transceivers, allowing remote control of devices via mobile applets, offering excellent performance and high cost-effectiveness, providing comprehensive solutions for intelligent speech products.
In 2023, the AI speech BLE chip CI231X series (CI2311, CI2312) was launched, integrating BNPU V3, with a system frequency of up to 220MHz, and built-in SRAM of up to 640KByte, providing more computing power for main control application solutions. It uses TSSOP24 packaging, with wider pins for easier production and lower patch costs, and integrates an RC oscillator, allowing for crystal-free applications after adding serial port adaptive functions. It supports speech recognition, voiceprint recognition, command word self-learning, voice activity detection, and deep learning noise reduction, with strong environmental noise suppression capabilities.
In 2022, the AI speech Wi-Fi Combo chip CI230X series (CI2305, CI2306) was launched, integrating Chipintelli’s self-developed brain neural network processor BNPU V3, as well as WiFi and BLE functions, creating a “Speech + WiFi + BLE” three-in-one AIoT chip, requiring only a few resistors and capacitors to implement various intelligent speech offline and online product hardware solutions, with very high cost-effectiveness. In terms of speech algorithms, these chips support speech recognition, single-microphone deep learning noise reduction, offline natural language processing, dual-microphone deep voice separation, command word self-learning version 2.0, and can support up to 500 command words. The CI2306 also supports voiceprint recognition.
In 2021, the AI speech chip CI13XX series (CI1301, CI1302, CI1303, CI1306, CI1311, CI1312) was launched. This series integrates BNPU V3, with a high main frequency (240MHz), larger SRAM, and stronger computing power. It is available in SSOP24/QFN40/SOP16 packaging, suitable for different applications and patch requirements. With higher integration and simplified peripherals, only resistors, capacitors, and PA chips are needed at the board level, aside from the MIC and speaker. It supports single-microphone noise reduction, AEC, dual-microphone enhancement, and dual-microphone directionality. Additionally, the CI1302, CI1303, and CI1306 support voiceprint recognition and offline natural speech.
In 2020, the AI speech chip CI112X series (CI1122) was launched, integrating BNPU V2, with lower cost and higher cost-effectiveness.
In 2019, the noise reduction and recognition integrated AI speech chip CI110X series (CI1102, CI1103) was launched, supporting offline far-field speech recognition, offline + online speech recognition, offline + AIoT Scene-based AIoT automation control, with richer integration (integrating noise reduction pickup, microphone array processing, ASR, MCU control, etc.), lower cost (at the same level as DSP and MCU chips), and lower power consumption (standby power consumption is less than one-tenth of AP chips with the same functional performance). It was evaluated as having an international advanced level by scientific and technological achievements (certificate number: 202051ZK3713) and was included in the “Selection Directory of AI Chip Technology 2020” initiated by AIIA.
In 2018, the AI speech chip CI1002 was launched, offering lower cost and wider application fields while maintaining the performance advantages of CI1006.
In 2017, Chipintelli launched the first domestic deep neural network AI speech chip CI1006, pioneering intelligent speech offline control of smart appliances, establishing Chipintelli’s leading position in the offline speech appliance field. It was evaluated as a domestic first by scientific and technological achievements (certificate number: 9562018Y0009) and was the only speech chip included in the book “The Wave of Artificial Intelligence.”
Algorithms¶
As a “Chip + Algorithm” company, Chipintelli possesses a full range of voice interaction technologies with completely independent intellectual property rights, including single-microphone and microphone array voice enhancement, noise suppression, echo cancellation, low-power wake-up, local semantic understanding, voiceprint recognition, and other speech algorithms. Our company have launched local self-learning technology, allowing users to customize voice commands, solving dialect recognition and user personalization challenges; our company have launched offline natural speech recognition technology, supporting tens of thousands of offline voice commands built into the chip, eliminating the need for users to memorize fixed voice commands, greatly enhancing the offline voice interaction experience. The speech recognition rate can reach over 97%, with a maximum recognition distance of 10 meters, leading the industry.
Solutions and Applications¶
Based on the company’s self-developed AI speech chips and AI algorithms, Chipintelli has launched a wide range of comprehensive product solutions applicable to smart homes, smart appliances, smart lighting, smart toys, smart robots, smart cars, and other fields. Our company has also opened the Chipintelli AI Speech Development Platform, aiming to help developers achieve more efficient and convenient intelligent speech product development.
After years of accumulation of AI voice and AI chip related technologies, our company currently leads the industry in the number of product solutions based on offline speech recognition chips, with thousands of enterprise partners developing millions of product projects based on the company’s chip products. We will continue to develop AI speech processing chip products and application solutions that better meet market and customer needs, aiming to accelerate the market penetration of intelligent speech products through continuous iteration of technology and products, making life better with intelligent speech technologies.
Company Mission¶
Lighting up life with AI