POLIPHONE
收藏资源简介:
POLIPHONE数据集是由米兰理工大学电子、信息与生物工程系创建的一个用于智能手机型号音频识别的数据集。该数据集包含来自20种不同智能手机的音频记录,涵盖语音、音乐和环境声音等多种领域,总时长超过50小时。数据集在受控环境中采集,确保了数据的一致性和可重复性,适用于多种实际应用场景。POLIPHONE数据集的创建旨在解决现有数据集过时或采集不一致的问题,确保数据集的可靠性和可扩展性,适用于多媒体取证和法律程序中的源归属问题。
The POLIPHONE dataset is a curated resource developed by the Department of Electronics, Information and Bioengineering at Politecnico di Milano for smartphone model audio recognition. It comprises audio recordings from 20 distinct smartphone models, spanning multiple domains including speech, music and ambient sounds, with a total duration of over 50 hours. Collected in a strictly controlled environment, the dataset guarantees data consistency and reproducibility, rendering it suitable for a wide range of real-world application scenarios. The POLIPHONE dataset was developed to address the shortcomings of existing datasets such as outdated content or inconsistent collection procedures, while ensuring its own reliability and scalability. It is applicable to source attribution tasks in multimedia forensics and legal proceedings.

- 1POLIPHONE: A Dataset for Smartphone Model Identification from Audio Recordings米兰理工大学电子、信息与生物工程系 · 2024年



