Researchers at the French National Centre for Scientific Research (CNRS) have determined how we form an impression of other people when we hear them speak for the first time. Our first judgments are made based on the rise and fall of the voice, which is also known as intonation.
People form mental images that represent the personality of another person according to the qualities of the individual’s voice. Now, for the first time, experts have developed models to visually express these mental representations.
The researchers created a computer program for voice manipulation called CLEESE. The software can assess the audio recording of a single word and randomly produce thousands of different pronunciations with unique notes.
The team analyzed the responses of study participants to find out which intonations made the word “hello” seem the most sincere and determined.
The researchers discovered, for example, a French speaker must pronounce bonjour with an emphasis on the second syllable in order to sound determined. On the other hand, the pitch must rise quickly at the end of the word to make the speaker sound sincere.
Using CLEESE, the team was able to visualize the “code” people use to judge others by their voices. The study revealed that the same code applied, regardless of gender.
The team members have already used their system to analyze how words are interpreted by stroke survivors, who often have an altered perception of intonation.
Ultimately, the researchers want to use their method to detect inconsistencies in how people perceive language and hope to use CLEESE as a tool for patient rehabilitation.
The study is published in the journal Proceedings of the National Academy of Sciences.