A Comparison of Human and Machine Estimation of Speaker Age
نویسندگان
چکیده
The estimation of the age of a speaker from his or her voice has both forensic and commercial applications. Previous studies have shown that human listeners are able to estimate the age of a speaker to within 10 years on average, while recent machine age estimation systems seem to show superior performance with average errors as low as 6 years. However the machine studies have used highly non-uniform test sets, for which knowledge of the age distribution offers considerable advantage to the system. In this study we compare human and machine performance on the same test data chosen to be uniformly distributed in age. We show that in this case human and machine accuracy is more similar with average errors of 9.8 and 8.6 years respectively, although if panels of listeners are consulted, human accuracy can be improved to a value closer to 7.5 years. Both human and machines have difficulty in accurately predicting the ages of older speakers.
منابع مشابه
A new classification method based on pairwise SVM for facial age estimation
This paper presents a practical algorithm for facial age estimation from frontal face image. Facial age estimation generally comprises two key steps including age image representation and age estimation. The anthropometric model used in this study includes computation of eighteen craniofacial ratios and a new accurate skin wrinkles analysis in the first step and a pairwise binary support vector...
متن کاملSpeaker Adaptation in Continuous Speech Recognition Using MLLR-Based MAP Estimation
A variety of methods are used for speaker adaptation in speech recognition. In some techniques, such as MAP estimation, only the models with available training data are updated. Hence, large amounts of training data are required in order to have significant recognition improvements. In some others, such as MLLR, where several general transformations are applied to model clusters, the results ar...
متن کاملSpeaker Adaptation in Continuous Speech Recognition Using MLLR-Based MAP Estimation
A variety of methods are used for speaker adaptation in speech recognition. In some techniques, such as MAP estimation, only the models with available training data are updated. Hence, large amounts of training data are required in order to have significant recognition improvements. In some others, such as MLLR, where several general transformations are applied to model clusters, the results ar...
متن کاملSome acoustic cues to human and machine estimation of speaker age
Two experiments were carried out in order to learn more about the relation between the various acoustic cues to speaker age. The first included listening tests with resynthesized stimuli, and the second comprised automatic estimation of age using the CART (Classification And Regression Trees) technique. In the first experiment, results indicate that human listeners seem to rely more on spectral...
متن کاملComparison and evaluation of intelligent models for river suspended sediment estimation (case study: Kakareza River, Iran)
Sediment transport constantly influences river and civil structures and the lack ofinformation about its exact amount makes management efforts less effective. Hence,achieving a proper procedure to estimate the sediment load in rivers is important. We usedsupport vector machine model to estimate the sediments of the Kakareza River in LorestanProvince and the results were compared with those obta...
متن کامل