This article proposes a comprehensive evaluation index, cost-effectiveness analysis model, adaptive encoding scheme, and improved denoising strategy for optimizing audio storage and noise removal. To improve audio quality and storage efficiency, and to quantify the balance between storage efficiency and sound quality fidelity for different audio formats, this article establishes a sound storage balance quantization model. By constructing a comprehensive evaluation quantization function and combining multidimensional factors to evaluate the balance between storage efficiency and sound quality fidelity. Based on a three-level indicator system (basic technology, performance evaluation, and comprehensive evaluation), entropy weight method is used to calculate indicator weights, and TOPSIS method is applied for comprehensive scoring. The results indicate that AAC format has the best sound storage balance performance, while WAV and FLAC have superior sound quality but lower storage efficiency, MP3 focuses more on storage efficiency in low bandwidth environments. At the same time, this article establishes a multidimensional audio evaluation model that quantifies sound quality using factors such as mean square error (MSE), while quantifying the size of audio files through factors such as audio duration, sampling rate, bit depth, number of channels, and compression ratio. Based on relevant features, the sound quality performance of speech and music content was evaluated, and the comprehensive and relative effects of sampling rate, bit depth, and compression algorithm on sound quality were analyzed. Based on this, an audio performance balance index was designed to quantify the cost-effectiveness between sound quality and file size. The sorting results show that music-44100Hz-MP3₃20kpp. mp3 and speech-44100Hz-MP3₃20kpp. mp3 have the highest cost-effectiveness, and recommend the best parameter configurations for voice and music audio.
Youmin Zhang (Thu,) studied this question.