Dialect separation assessment using log-likelihood score distributions
نویسندگان
چکیده
Dialect differences within a given language represent major challenges for sustained speech system performance. For speech recognition, little if any knowledge exists on differences between dialects (e.g. vocabulary, grammar, prosody, etc.). Effective dialect classification can contribute to improved ASR, speaker ID, and spoken document retrieval. This study, presents an approach to establish a metric to estimate the separation between dialects, and to provide some sense of expected speech system performance. The proposed approach compares dialects based on their loglikelihood score distributions. From the score distributions, a numerical measure is obtained to assess the separation between resulting GMM dialect models. The proposed scheme is evaluated on a corpus of Arabic dialects. The sensitivity of the dialect separation score is also quantified based on controlled mixing of dialect data for the case of measuring dialect training data purity. The resulting scheme is shown to be effective in measuring dialect distance, and represents an important objective way of assessing dialect differences within a common language.
منابع مشابه
Hyperbolic Cosine Log-Logistic Distribution and Estimation of Its Parameters by Using Maximum Likelihood Bayesian and Bootstrap Methods
In this paper, a new probability distribution, based on the family of hyperbolic cosine distributions is proposed and its various statistical and reliability characteristics are investigated. The new category of HCF distributions is obtained by combining a baseline F distribution with the hyperbolic cosine function. Based on the base log-logistics distribution, we introduce a new di...
متن کاملAutomatic Mandarin pronunciation scoring for native learners with dialect accent
This paper studies pronunciation scoring algorithm in CALL system aiming at teaching native Chinese learn standard Mandarin. Most of the pronunciation scoring algorithms focus on non-native environment, which may not be suitable for native speakers. We bring up a new algorithm based on traditional posterior log-likelihood algorithm by weighting the initial part of Mandarin syllables, where fina...
متن کاملThe distribution of calibrated likelihood-ratios in speaker recognition
This paper studies properties of the score distributions of calibrated log-likelihood-ratios that are used in automatic speaker recognition. We derive the essential condition for calibration that the log likelihood ratio of the log-likelihood-ratio is the log-likelihood-ratio. We then investigate what the consequence of this condition is to the probability density functions (PDFs) of the loglik...
متن کاملSource separation in post nonlinear mixtures: an entropy-based algorithm
This paper proposes a new approach for sources separation in special nonlinear mixtures, called post nonlinear mixtures (PNL). We rst explain the nice separability properties of these mixtures: solutions have almost the same indeterminacies as in instantaneous linear mixtures. The method proposed in this paper is based on the minimization of the mutual information, which needs the knowledge of ...
متن کاملBlind Source Separation Using Maximum Entropy Pdf Estimation Based on Fractional Moments
Recovering a set of independent sources which are linearly mixed is the main task of the blind source separation. Utilizing different methods such as infomax principle, mutual information and maximum likelihood leads to simple iterative procedures such as natural gradient algorithms. These algorithms depend on a nonlinear function (known as score or activation function) of source distributions....
متن کامل