There has been much concern in Europe with regard to the correct logic for the evaluation of evidence whereas in the United States demonstration of validity and reliability has been the primary concern. Among those who are uneasy with, or outright oppose, the use of likelihood ratios there is concern as to the basis on which values are assigned to likelihoods. This has been expressed as a concern regarding the warrant for the strength of evidence expressed by the forensic scientist - Where do the numbers come from? A paradigm shift is underway in forensic science, and we have formulated a description of the new paradigm which addresses the major concerns on both sides of the Atlantic and the abovementioned concern of the sceptics. In our formulation the new paradigm includes the following key elements: *Use of the likelihood ratio framework as the logically correct framework for the evaluation of forensic evidence, including consideration of what the relevant hypotheses are and by extension what the relevant population is in the particular case under investigation. *Use of relevant data, quantitative measurements, and statistical models to calculate numeric likelihood ratios, and presentation of the numeric values as strength of evidence statements. *Empirical testing of the degree of validity and reliability of the forensic analysis system under conditions reflecting those of the particular case under investigation. *Transparency as to decisions made and procedures adopted as part of the forensic analysis. In this presentation we describe our formulation of the new paradigm and how it addresses current concerns. We also outline a concrete example based on a particular real forensic voice comparison case, which is representative of a common type of case.