Accelerate the in-vehicle digital experience with Azure Cognitive Services | Azure Blog and Updates

[ad_1]

Microsoft helps to reshape the automotive trade in the way in which it serves its drivers with in-vehicle infotainment programs. For example, Azure is partnering with XPeng to allow AI voice experiences for automotive manufacturers and prospects. The answer gives the trade with a contemporary tackle text-to-speech and expressive voice, world languages, speaker constancy, and self-service customization. XPeng joins a rising pattern of automakers rethinking investments in environmental voice.

“It is a cutting-edge exploration of auto voice interplay within the auto trade,” Xpeng automotive AI product senior skilled Hao Chao mentioned. “The expertise delivers an entire new degree of pure speech. With a deep understanding of city mobility, we’re discovering many extra eventualities to leverage AI expertise for a excessive degree of driver-machine instinct.”

XPeng tapped into Microsoft’s neural text-to-speech expertise for his or her in-car consumer expertise. By utilizing Microsoft’s neural text-to-speech with emotional types, Xpeng can present a extra pleasant listening expertise for his or her prospects and fight listening fatigue. Microsoft’s neural text-to-speech gives fluency and naturalness that’s similar to a human voice. Coupled with multi-emotional voices, Microsoft text-to-speech acts as a refreshing substitute to the monotonous sound many automotive assistants have right this moment.

“We’re excited to reimagine how speech and voice can enhance the lives of drivers,” Azure AI Speech Product Lead Binggong Ding mentioned. “Whereas from a technical viewpoint, we actually wish to make this a mannequin that may serve all auto manufacturers and their builders. How can we greatest optimize using artificial speech to allow a high-fidelity voice expertise with out compromising sound high quality? XPeng is constructing upon this problem to offer a voice assistant that prospects have been on the lookout for.”

Microsoft’s long-term purpose is to make superior multi-emotional, world voice capabilities the brand new customary for world automotive manufacturers and customers. The expertise adopted by XPeng added dozens of voice types, distinctive emotional depth management, and deduction skills. It covers 90 certifications worldwide together with home insurance policies, regulatory knowledge middle requirement and EU GDPR, and better knowledge privacy-policy holder necessities. Along with the automotive producers, Microsoft is creating new driving experiences with speech primarily based on the text-to-speech and speech-to-text capabilities inside Azure Cognitive Providers for speech.

Accelerated speech innovation

Voice is the brand new interface in ambient computing expertise. The standard of text-to-speech and speech-to-text has improved lately as a consequence of analysis and technological leaps enabled by the event of neural networks. Excessive-quality speech-to-text and text-to-speech fulfill the wants of the automaker to create the following era fashionable in-car speech expertise. Microsoft speech-to-text affords strong recognition capabilities that are speaker-independent and able to dealing with ambient noise whereas driving. Microsoft text-to-speech additionally includes a extra fluid, natural-sounding voice which could be a differentiation for automakers and prospects alike. Each speech-to-text and text-to-speech additionally improve hands-free management of the automotive infotainment system. Microsoft text-to-speech helps a number of talking types, together with chat, newscast, and customer support. These developments permit drivers to have a extra pleasant driving expertise. For extra details about the current developments in speech-to-text and text-to-speech take a look at speech-to-text with its analysis outcomes, reaching human parity on the Switchboard analysis benchmark and neural-text-to-speech is near human-parity.

Providing world languages

Microsoft helps automakers cowl their world enterprise and only recently hit a milestone of 100 languages and now helps 119 languages and variants with 278 voices out-of-box. That is aligned with our firm imaginative and prescient to empower each particular person and group on the planet to realize extra. “100 languages is an effective milestone for us to realize our ambition for everybody to have the ability to talk whatever the language they converse,” mentioned Xuedong Huang, Microsoft Technical Fellow and Azure AI Chief Expertise Officer. With extra languages with their variants coated, we’re excited to be powering pure and intuitive voice experiences for automakers.

Differentiation with customization

Microsoft empowers automakers to develop a extremely life like branded voice for extra pure conversational interfaces utilizing the customized neural voice functionality. Based mostly on the neural text-to-speech expertise and the multi-lingual multi-speaker common mannequin, customized neural voice allows you to create artificial voices which are wealthy in talking types or adaptable cross languages with as little as 30 minutes of audio. The life like and natural-sounding voice of customized neural voice can symbolize manufacturers and particular personas and permit customers to work together with purposes naturally in a conversational type. Try this weblog for a step-by-step information on methods to create a customized neural voice.

Compliance and accountable AI

Microsoft is dedicated to investing in assembly regulatory requirements across the globe to satisfy the automakers’ compliance necessities. The speech service, a part of Azure Cognitive Providers, is licensed by SOC, FedRAMP, PCI DSS, HIPAA, HITECH, and ISO. Backed by Azure infrastructure, the speech service additionally affords enterprise-grade safety, availability, compliance, and manageability.

Microsoft is dedicated to growing AI expertise in a accountable means. We use completely different technical and coverage options to safeguard in opposition to misuse of the expertise. For instance, we’re designing and releasing Customized Neural Voice with the intention of defending the rights of people and society, fostering clear human-computer interplay, and counteracting the proliferation of dangerous deepfakes and deceptive content material. This aligns with Microsoft’s dedication to accountable AI. That dedication contains Transparency Notes, which communicates the aim, capabilities, and limitations of an AI system.