The Tone of Your Voice May Have an Effect on Your Marital Relationship

The way spouses speak to one another may predict if their marriage is ultimately a success or failure, suggests a paper that was published in Proceedings of Interspeech.

Researchers said that marital results have always intrigued many clinical psychologists. The researchers are in the process of gathering information from couples' speech, visual gestures, and spoken language in order to keep an eye on these key points, understand, and even anticipate the end result of their relationships. The information they're gathering are vital information, but these researchers believe that spoken interactions may be the most definitive among the many.

In their written report, they examined the importance of various acoustic features taken from the couples' respective tone of voice, as well as whether these can provide information such as relationship satisfaction.

Researchers examined recordings of 134 real couples, who have been married for 10 years during problem-solving interactions. These recordings and coded sessions happened before therapy began, and again after 26 weeks, and the last is the second year of the therapy. The sessions last for 10 minutes and had 2 sub-sessions. Before the sessions started, the husband and wife were made to think of a topic respectively and then discussed it during the sub-sessions.

The couples' interactions were rated on a scale of one to four, with one being the lowest which means the relationship has deteriorated, and believed to have gotten worse over the course of the treatment. Four on the other hand is the highest which indicates that the relationship has recovered and that it got better over the course of the treatment.

The researchers then applied a method they have developed to break the recordings down into its acoustic features using speech-processing techniques. The feature measured everything, from the loudness and its by-product, to intensity, pitch, and a certain characteristic researchers call "jitter" and "shimmer."

Researchers guessed that the features can capture useful information about the mutual and self-influence of behavioral patterns of the speakers over a period of time, and they were right.

Real Time Analytics