Spotify is developing speech-recognition tech to detect your mood and suggest music


Global music-streaming giant Spotify has recently been granted a patent for a new technology which will allow it to analyze its users' voice data – including speech recognition and background noise – with an eye to suggesting music based on what it finds.

As reported by Music Business Worldwide, this patent was first applied for in February 2018 and was granted January 2021. The filing states that Spotify's intention is to use audio recognition to identify traits such as “an emotional state, a gender, an age, or an accent of the speaker” and then recommend content based on this model.

This isn’t the first time the Swedish company has toyed with what some might view as 'overly personal' personalization techniques – a previous patent that Spotify had granted involved customizing a user’s experience based on their ‘personality traits’.

In both cases, the aim for Spotify is to target users with more relevant and appropriate content, whether it be music or podcast recommendations, or potentially advertising.

The company is known for its algorithm that suggests new audio content for its users based on previous listening habits, so it’s unsurprising that it would want to add other metrics to enhance its capabilities and accuracy.

Privacy vs personalization

As giant tech companies continue to grow and optimize their platforms for both users and advertisers, personal data privacy is a constant topic of public concern, Facebook’s Cambridge data scandal was a particularly well-documented example from 2018, but there have been a plethora of other examples of consumer data being misused, from fitness tracking companies to adult streaming sites.

With the rise of smart speakers came the concern that an ever-listening device would, again, breach the privacy of its users by recording information in an unsolicited fashion, and using it as a data point in a tech company’s mass analytics.

The continuing popularity of Google's Home and Nest speakers and Amazon's Echoes (not to mention the integration of both company's voice-assistants across a wide range of third-party devices) would seem to indicate that most of us are happy with a certain level of potential intrusion on privacy for the sake of convenience – but it’s fairly unanimous that consent plays an important factor in this balance.

While Spotify listeners may embrace and enjoy the algorithmic personalization that the service provides, it’s possible that some of these more deeply personal (and potentially invasive) methods for collecting the data won’t be accepted quite so easily.



from TechRadar - All the latest technology news https://ift.tt/39tbFm8

No comments