dating AI can predict whether your relationship will last based on how you speak to your partner

[captain_to_fire]

Source: PLOS One via Phys.org

journal.pone.0185123.pdf

 

Nasir, M., Baucom, B. R., Georgiou, P., & Narayanan, S. (2017, September 21). Predicting couple therapy outcomes based on speech acoustic features. (I. McLoughin, Ed.) PLOS One, 23. doi:https://doi.org/10.1371/journal.pone.0185123

Quote

ABSTRACT

Automated assessment and prediction of marital outcome in couples therapy is a challenging task but promises to be a potentially useful tool for clinical psychologists. Computational approaches for inferring therapy outcomes using observable behavioral information obtained from conversations between spouses offer objective means for understanding relationship dynamics. In this work, we explore whether the acoustics of the spoken interactions of clinically distressed spouses provide information towards assessment of therapy outcomes. The therapy outcome prediction task in this work includes detecting whether there was a relationship improvement or not (posed as a binary classification) as well as discerning varying levels of improvement or decline in the relationship status (posed as a multiclass recognition task). We use each interlocutor’s acoustic speech signal characteristics such as vocal intonation and intensity, both independently and in relation to one another, as cues for predicting the therapy outcome. We also compare prediction performance with one obtained via standardized behavioral codes characterizing the relationship dynamics provided by human experts as features for automated classification. Our experiments, using data from a longitudinal clinical study of couples in distressed relations, showed that predictions of relationship outcomes obtained directly from vocal acoustics are comparable or superior to those obtained using human-rated behavioral codes as prediction features. In addition, combining direct signal-derived features with manually coded behavioral features improved the prediction performance in most cases, indicating the complementarity of relevant information captured by humans and machine algorithms. Additionally, considering the vocal properties of the interlocutors in relation to one another, rather than in isolation, showed to be important for improving the automatic prediction. This finding supports the notion that behavioral outcome, like many other behavioral aspects, is closely related to the dynamics and mutual influence of the interlocutors during their interaction and their resulting behavioral patterns.

CONCLUSION

In this article, we presented a study on automatically predicting the marital relationship status of distressed couples in therapy using acoustic information from their speech. We presented a framework for capturing behaviorally significant acoustic features from the spoken interactions of couples engaged in problem solving discussions. We also introduced knowledge-driven features of capturing short-term and long-term acoustic descriptors inspired by previous studies on human interactions. We compared this automatic approach of capturing important behavioral information directly from speech signal to the traditional approach taken by psychologists, i.e., manual coding of behavior from therapy sessions.

In the multiple classification experiments, we observed that the acoustic features from speech capture more relevant information than the manually constructed behavioral dimensions for predicting the marital outcomes from human experts. This is a promising finding considering the fact that human coders had utilized multiple modalities (speech, visual and lexical information) in their coding process. Even though behavioral codes are not designed to predict outcomes itself, they function as behavioral descriptors of the couple and one can expect them to be informative towards the outcome based on the observational methods of psychology.

We also found that dynamic functionals are better than traditional static functionals of acoustic features for outcome prediction. This work opens up avenues for many other research applications and similar frameworks for various behavioral outcome prediction tasks such as assessing results of treatment for various disorders and conditions.

In the future, we can also analyze the importance of other communication modalities including language use (i.e., what is being spoken), and visual (e.g., head-movement and other face and body expressions). One can also investigate more complex temporal modeling (e.g., hidden Markov models, dynamical systems modeling) of the behaviors captured through the acoustic features extracted from the speech signal. Also, automatic recognition of the mental states (such as emotional arousal) of the speakers and investigation of the dynamics of local behavioral cues might be useful. (Emphasis is mine)

I think Taylor Swift needs this as she makes songs based on the long list of her ex-boyfriends while writing songs about them and probably she'll stop playing victim. 

Quote

New research, just published in the journal PLOS ONE, has analysed the vocal characteristics of 134 couples undergoing therapy. Researchers from the University of Southern California used computers to extract standard speech analysis features from recordings of therapy session participants over two years. The features – including pitch, variation in pitch and intonation – all relate to voice aspects like tone and intensity.

 

A machine-learning algorithm was then trained to learn a relationship between those vocal features and the eventual outcome of therapy. This wasn't as simple as detecting shouting or raised voices – it included the interplay of conversation, who spoke when and for how long as well as the sound of the voices. It turned out that ignoring what was being said and considering only these patterns of speaking was sufficient to predict whether or not couples would stay together. This was purely data driven, so it didn't relate outcomes to specific voice attributes.

 

Interestingly, the full video recordings of the therapy session were then given to experts to classify. Unlike the AI, they made their predictions using psychological assessment based on the vocal (and other) attributes – including the words spoken and body language. Surprisingly, their prediction of the eventual outcome (they were correct in 75.6% of the cases) was inferior to predictions made by the AI based only on vocal characteristics (79.3%). Clearly there are elements encoded in the way we speak that not even experts are aware of. But the best results came from combining the automated assessment with the experts' assessment (79.6% correct).

 

The significance of this is not so much about involving AI in marriage counselling or getting couples to speak more nicely to each other (however meritorious that would be). The significance is revealing how much information about our underlying feelings is encoded in the way we speak – some of it completely unknown to us. (Emphasis is mine)

 https://phys.org/news/2017-09-ai-relationship-based-partner.html#

I can see two things happening here if this AI, rising divorce rates because an AI says they are incompatible or marriage counselors out of their jobs because couples put their trust on an Ai kinda like how a lot of people self-diagnose and treat their diseases on the internet. I am not a relationship or a dating expert and I have no plans of becoming like Taylor Swift who dated at least a dozen people. But therapists can use them as tool to help them during couples counselling. I can see app developers like Tinder, Ok Cupid and others jumping in to do self-help apps like placing the phone in the middle of a table while talking and the app will analyze if the other person is honest or truly in love or not. But I'm not yet convinced that analyzing pitch and tone alone of someone's voice while talking to their partner is a best way to determine if a marriage is doomed to fail or it is still worth saving. I still think that a relationship especially a marriage is based on communication and trust. I think a lot of relationships fail either because the other person is a philandering piece of shit or both the people involved didn't explored and tested the waters enough to determine if they're actually compatible or one or both people in the relationship change and either one of them is not a fan of it.

[FPSwithaWacomTablet]

Ooo.. Well.. Umm..

 

There goes all my years of dating sims gone to waste. That AI played it all.

 

/s

[Ryujin2003]

"This just in: Talking to your significant other like a Douchebag may be detrimental to your relationship. More on this at 11."

[captain_to_fire]

Just now, VegetableStu said:

Oh hey "AI does X from datasets generated from humans/human input" news. Now to put on thinking hat to figure out how to troll it.

I can see Amazon adding this kind of AI to Alexa in their Echo products. "Alexa, is my wife cheating with my best friend?"

[TidaLWaveZ]

BEEP - I GIVE EM TWO WEEKS (robot voice)

[ATFink]

52 minutes ago, VegetableStu said:

"Here are past purchases sent to Randy"

"THAT BIC- I'M GOING TO KILL HER"

"Playing 'Murder' by Diana Glory"

Yes, but which voice is Alexa's? 

Maybe we service the robots in the future! (and have really good vocals)

[Sauron]

I suppose statistically it works, but I wouldn't base myself on statistics for something like this.

[Ryujin2003]

1 hour ago, hey_yo_ said:

I can see Amazon adding this kind of AI to Alexa in their Echo products. "Alexa, is my wife cheating with my best friend?"

Come home, your wife left the milk on the counter for the millionth time.. then you say to yourself with, I'm gonna kill her.

Alexa calls the police for domestic violence and find your wife a divorce lawyer.

[captain_to_fire]

13 minutes ago, Sauron said:

I suppose statistically it works, but I wouldn't base myself on statistics for something like this.

I think for an AI like this or any other form of machine learning, it needs to have a big set of voice data for it to determine if the relationship is doomed or not based on their voices while talking. What if the AI said that they need to have a divorce when in fact they just need more intimate time together.

[Silentprototipe]

Why am I not surprised

[Okjoek]

Computers breeding the perfect human race? Sounds kind of cool to me.

[SCHISCHKA]

I did not read the original research I'm going by the face of it. It reminds me of the onion video with title "research shows stabbing may be harmful to monkeys". They are trying to make a model that predicts relationship length based on how couples talk? If this is right then that is pretty fucking obvious. Is it worth going deeper into seeing what this models is?

[captain_to_fire]

19 minutes ago, SCHISCHKA said:

I did not read the original research

I attached the PDF file in the OP. 

20 minutes ago, SCHISCHKA said:

They are trying to make a model that predicts relationship length based on how couples talk?

Yup. And surprisingly, the results of the AI is close to the conclusions by the therapists. 

[captain_to_fire]

30 minutes ago, Okjoek said:

Computers breeding the perfect human race? Sounds kind of cool to me.

Not really but more of determining the perfect relationship. If you want stuff about designer babies and making humans close to perfection. 

 

[samcool55]

Hmm afaik girls stay with douchebags even tho they are douchebags so i'm wondering if this is right.

Also depending where you are it will have a damn hard time figuring out if people are talking to their SO or a family member

[mynameisjuan]

1 hour ago, samcool55 said:

Hmm afaik girls stay with douchebags even tho they are douchebags so i'm wondering if this is right.

Sad to say how true this actually is......

[FRSHPRNCFBLR]

So...the real question is: Will the AI think that you're in a relationship with your father if she calls you daddy.

[Tech_Dreamer]

AI Preview.

[GOTSpectrum]

Interesting, but I don't think anyone is going to be like, 'hey, I think we should break up because Alexa says so'......

on the other hand.... 

[captain_to_fire]

2 hours ago, Ben Quigley said:

Interesting, but I don't think anyone is going to be like, 'hey, I think we should break up because Alexa says so'......

on the other hand.... 

If some people are stupid enough to self diagnose a stomach ache or strep throat on the internet, some people will be shallow enough to break up a relationship simply based on an AI. 

 

[captain_to_fire]

2 minutes ago, VegetableStu said:

not sure if facepalm or pity ._.

It terrifies me a bot in the future. Not only we’ll be replacing jobs with machines but we’re now putting intimacy and communication issues of couples with a machine. 

 

“Hey Alexa, why is my husband so cold in our last intimate moment?”

”It’s because he’s cheating on you with your daughter’s ballet teacher.” 

 

You see how shallow it is. What if the husband just needs some blue pills? 

[bigneo]

wtf is this advanced AI...

 

I've seen an AD on TV where you can text yours and partners names only and it will tell you if it will last or not. lel

[captain_to_fire]

1 minute ago, bigneo said:

wtf is this advanced AI...

 

I've seen an AD on TV where you can text yours and partners names only and it will tell you if it will last or not. lel

What ad is that? Do you have a YouTube video?

[bigneo]

2 minutes ago, hey_yo_ said:

What ad is that? Do you have a YouTube video?

it was a joke  but I'll try to find it. 

 

Pretty much the ad was to get your money by sending text to a given number