Suzuki, A., Watanabe, S., Taheno, Y., Kosugi, T., & Kasuya, T. (1973). Possibility of detecting deception by voice analysis.  Reports of the National Institute of Police Science, 26, 62-66.

The present study examined characteristics of voice pitch, intensity and duration that might be related to the act of deception.  If the act of deception produces some involuntary changes in voice, there should be some differences between deceptive responses and non-deceptive responses.  Suzuki et al. (1973) analyzed tape-recording of deceptive responses from 3 people who had been confirmed as guilty of criminal acts.  Deceptive responses were compared to non-deceptive responses in the analysis of voice pitch, intensity and duration.  Voice pitch and intensity showed no changes that distinguished deceptive responses from non-deceptive responses.  The accuracy rate in the detection of deception was lower than the chance level for both pitch and intensity.  For voice duration, deceptive responses showed longer duration than non-deceptive responses, but the difference was not statistically significant.  Moreover, The accuracy rate did not exceed the chance level.  Thus, it was concluded that there was no evidence that the voice analysis provides information to detect the act of deception.