On January 7, 2022, Microsoft officially announced that with deep neural Network TTS (text-to-Speech) support based on Microsoft Smart Cloud Azure,Xiaopeng Automobile, A leading smart electric vehicle company in China, has successfully completed the upgrade of its vehicle-level voice assistant, further upgrading the technical level of intelligent vehicle voice assistant.
Currently, Chinese buyers of Xiaopeng's P7 model can upgrade its new voice-friendly intelligent assistant, The Little P, via over-the-air (OVER-THE-air), and xiaopeng plans to introduce The technology to several other models via OTA.
Thanks to Microsoft's work in speech, natural language and machine translation over the past few years, the technology has improved dramatically in terms of fluency, quality, fidelity and naturalness.
These innovations, combined with Microsoft's Azure AI technology and other products, have helped companies like Xiaopeng bring richer and more attractive user experiences to their consumers.
In the months of cooperation, Microsoft and Xiaopeng Automobile have worked together to overcome three technical challenges facing the application of speech synthesis technology:
First of all, in order to solve the network jitter problem in automobile scenarios and ensure the continuous operation of the voice function with high quality, Xiaopeng automobile built a multi-level cache architecture, which can preset and cache high-quality voice files in advance, reducing the function's dependence on the network to the greatest extent.
Second, under the premise of resources in order to do not take up too much, still can provide comparable to the real voice of human experience, xiao peng car with the help of the Microsoft smartphone Azure cloud caching and compression, audio files can be compressed into 24 KHZ sampling rate and the quantitative level of 16, greatly reduces the data network and the work force resources pressure;
Finally, many improvements have been made to reduce the ambiguity of synthesized speech and optimize the accuracy of polyphonic words.
Thanks to the efforts of both parties, the new on-board voice synthesis function has reached a new level in terms of voice fidelity, functionality and scene optimization, and Xiaopeng is able to deploy the voice assistant in more usage scenarios, making it an integral part of the intuitive driving experience.
Hao Chao, senior expert of AI products of Xiaopeng Automobile, said: "From the determination of cooperation intention to the launch of the product, we have spent several months with Microsoft to jointly complete a cutting-edge exploration of automobile voice interaction technology, which has raised the natural voice level of vehicle voice to a new level. "As the understanding of urban mobility deepens and more scenarios are explored, these technologies will be widely applied to achieve a high-level human-machine co-driving experience."
"As research and technology advances, Azure cognitive services such as vision and voice will play a key role in defining unique in-car experiences," said Sanjay Ravi, General manager of Microsoft's Automotive, Mobility and Transportation industry. "Intelligent voice is emerging as a major in-vehicle interaction tool, and Microsoft's pre-built deep neural voice and personalized deep neural voice customization will help automakers strengthen their brands and create differentiated and authentic user experiences that are closer to natural human voices."
In addition to Xiaopeng, Microsoft has also carried out in-depth cooperation with a number of automobile manufacturers and partners in the field of intelligent vehicles, focusing on promoting the intelligent application of the automotive industry.
Different manufacturers have different intelligent needs. From human-computer interaction to driving information analysis, judgment and decision, different brands and vehicles need to load intelligent applications with different needs.
Based on the underlying platform of powerful voice semantics and data architecture, Microsoft empowers many intelligent automobile manufacturers with powerful technical capabilities and the underlying platform, and develops the central control display voice system of various information and data, as well as multi-dimensional hardware structure, so that users can experience more intelligent cockpit interaction feelings.