语音相似度评价算法分析.docx

Research on ic similarity evaluation algorithm
Specialty: Signal and Information Processing Name: Ren Xue Ni
Instructor: Professor Wang Min
ABSTRACT
With the improvement of the people's material and spiritual culture more and more abundant, many singing contest program ratings link is an integral part. But Most of the scoring method is the authority of the judges in the field to decide the basis of their professional skills, as well as inside and outside audience support players’ performance results. The results of this rating will inevitably mixed personal subjective factors of the judges and the audience, there is a dispute arising due to personal preferences. Existing scoring software may be due to the unreasonable rated parameters and limited similarity matching, made the final score is not very satisfactory. In order to solve the problem, looking for a quick and effective voice evaluation method is particularly important.
Voice similarity evaluation is an automatic evaluation method for the degree of similarity voice of original singers and imitators, Its process is to extract feature extraction and similarity matching from input standard voice and the imitate voice. Make puter simulation its Subjective impression for similarity to evaluate.
To ensure the accuracy of speech recognition voice is the premise for an accurate assessment. Voice pre-emphasis, a windowing function, framing a series of pre-treatment is the first; the purpose is to reduce the background noise in the voice, to improve the purity of the voice . Use the method bining short-time average energy and short-time average zero-crossing rate for voice endpoint detection feature extraction, then sequence the characteristic parameters into pattern matching, and came to the conclusion rated. Characteristic parameter extraction and similarity matching is the core of the voice similarity evaluation system. In Feature extraction, select the pitch frequency, Mel cestrum coefficient and the intensit