On January 7, 2022, Microsoft officially announced that with the support of TTS (Text-to-Speech), a deep neural network based on Microsoft's intelligent cloud Azure, Xiaopeng Automobile has successfully completed the upgrade of its vehicle-level voice assistant, further improving the technical level of intelligent car voice assistants.
As voice becomes the primary interactive tool within the car, Microsoft's custom neural voice service enables automakers to develop their own differentiated and authentic brand experiences. Microsoft's research breakthroughs in speech, natural language, and machine translation have helped dramatically advance the fluency, quality, fidelity, and naturalness of voice assistant technology over the past few years. These innovations have been integrated into the commercial speech and language capabilities of Azure Cognitive Services and other Microsoft products, so companies like Xiaopeng Motors can bring richer and more engaging experiences to their customers.
Working with Microsoft, Xiaopeng Motors overcame several key challenges to create new cutting-edge voice assistant integrations. In order to cope with network jitter during the driving of the car, while reducing data traffic consumption and hardware burden, and ensuring continuous high-quality voice, Xiaopeng Motors has introduced multi-level caching for specific environments, caching high-quality sound in advance to minimize dependence on the network.
To deliver natural-sounding, high-fidelity voice, Xiaopeng Motors leverages Microsoft Azure's caching and compression capabilities to deliver a high-quality voice sample rate of 24KHz and a quantization level of 16Bit without placing an undue burden on the data network or the car's own CPU. Xiaopeng Motors has also partnered with Microsoft to minimize ambiguity and optimize the accuracy of voice assistant speech.
As a result, the new voice assistant features reach new levels of realistic speech fidelity, functionality, and scene-specific applicability. With these new features, Xiaopeng Motors can deploy voice assistants in more use cases, making voice assistants an integral part of the intuitive driving experience.
"This is the automotive industry's cutting-edge exploration of vehicle voice interaction," said Hao Chao, a senior expert on Xiaopeng Motors' AI products. "It took months of dedicated work on our team to overcome the challenge and now offers a whole new level of natural voice." With an in-depth understanding of urban mobility, we find that there are more scenarios where AI technology can be leveraged to achieve a high level of driver-machine intuition."
"As research and technology advance, Azure cognitive services such as vision and speech will play a key role in defining unique in-vehicle experiences," said Sanjay Ravi, general manager of Microsoft's automotive mobility and transportation industry. "Because voice is the primary interactive tool in the car, Microsoft's custom neural voice service enables automakers to develop their own differentiated and authentic brand experiences."
It is reported that Chinese users who purchase the P7 model of Xiaopeng Motors can upgrade this new intelligent voice assistant "Xiao P" that can be comparable to the beautiful voice through OTA, and Xiaopeng Motors also plans to introduce this technology upgrade into other mass production models through OTA.