 |
| Spotify |
Spotify, a global music streaming giant, has recently been granted a patent for a new technology that will allow it to analyze the voice data of its users, including speech recognition and background noises based on the emotional state, gender, age, or an accent of the speaker with a view to recommending music based on what it finds.
First applied for in February 2018 and granted on 12 January 2021, the patent that will allow more detailed music recommendations enables the streaming giant to "make observations" by recognizing features of the environment and emotions of a user using speech recognition technology and then recommending content based on this model.
According to the patent, Spotify will then play music representing their mood or even their social atmosphere, according to the patent - "e.g. alone, small group, party."
It would potentially raise the streaming numbers for Lesley Gore's It's My Party (And I'll Cry If I Want To) if the feature is introduced.
Spotify stressed that in its application, "It is common for a media streaming application to include features that provide personalized media recommendations to a user". It is common for a media streaming application to include features that provide a user with personalized media recommendations.
The current approach to discerning the preferences of a consumer is unsatisfactory, however, since it needs them to "to tediously input answers to multiple queries" regarding their age, gender and favorite bands.
This is not the first time that the Swedish company has used a previous patent that Spotify had issued to customize the experience of a user based on their 'personality characteristics' through what others may see as 'overly personal' personalization techniques.
Spotify said its mission is to target users with more relevant and acceptable content, whether it is recommendations for music or podcasts or possibly advertising.
Based on past listening patterns, the company is known for its algorithm that recommends new audio material for its customers, so it's unsurprising that it would want to add other metrics to boost its capabilities and accuracy.
Karaoke feature
The patent suggests that speech recognition may be used to collect age and gender information, while contextual indicators such as "intonation, stress, [and] rhythm" can provide clues as to whether a consumer is "happy, angry, sad or neutral".
"It should be understood that the above example metadata categories of emotions, gender, age and accent are merely examples, and numerous other characterizations and classifications can be used," it said in the filing.
To enhance recommendations, the findings will be coupled with other knowledge - like the previously played songs of a user, and the taste of their friends in music.
The filing demonstrates how speech recognition can also be used to enhance your playlists
One of the variables taken into account will be speech recognition.
The technology is one of several technological advancements patented over the past year by Spotify.
It patented a karaoke-like feature in September that enables individuals to "overlay a music track with their own vocals" And it received permission earlier this week for a "cadence-based media content selection engine" that matches your music's tempo to your running pace.
It also conducted a study last year on how the character characteristics of people affect their musical tastes.
Spotify's Research and Development team concluded that "Those who scored high in conscientiousness tended to concentrate their listening to a more narrow window of time of day."
On the other hand, self-assessed introverts appeared to dig deeper into the repertoire of an artist, listening to more tracks for each artist they found.'
With 320m users, including 144m subscribers, Spotify is the world's largest audio streaming service.
A patent has been granted to the streaming service, which has 320 m global users, that would allow it to use such insights to promote content that has been tweaked to appeal to the personality of a listener, including music, podcasts and ads.
"In some embodiments in which the personalized content includes one or more messages with audio components," the patent states, "the electronic device changes a tone of voice for messages for presentation to the user."
"For example, the tone of voice may be more upbeat, high-pitched and/or exciting for users that have been assigned the personality trait of extroversion."
Although such tracking technology seems dystopian, without due regard to the ethical consequences, Spotify's own researchers have cautioned against implementing it.
They added that "We recognize that one's digital history is extraordinarily personal and sensitive," As such, from its access, it must be handled with careful consideration of possible misuse.
"We disavow any future research or applications that violate ethical standards of data usage and are not transparent about privacy to its users".
Privacy vs personalization
As giant tech firms continue to expand and refine their services for consumers and marketers, privacy of personal data is a constant public problem, the Cambridge data scandal on Facebook was an especially well-documented example from 2018, but there have been a plethora of other instances of abuse of customer data, from fitness monitoring companies to adult streaming sites.
With the advent of smart speakers, the fear emerged that, once again, an ever-listening system would violate the privacy of its users by capturing information in an unsolicited way and using it as a data point in the mass analytics of a tech company.
The continued success of Google's Home and Nest speakers and Amazon's Echoes (not to mention the integration of the voice assistants of both companies into a wide variety of third-party devices) would seem to suggest that most of us are content for convenience with a certain degree of potential privacy infringement, but it is very unanimous that consent plays an important role in this balance.
Although the algorithmic personalization that the service offers can be welcomed and appreciated by Spotify listeners, it is likely that some of these more intensely personal (and potentially invasive) data collection approaches will not be approved quite so easily.
No comments:
Post a Comment