{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/cleanmel-mel-spectrogram-enhancement-for","title":"CleanMel: Mel-Spectrogram Enhancement for Improving Both Speech Quality and ASR","arxiv_id":"2502.20040","date":"2025-02-27","proceeding":null,"authors":["Nian Shao","Rui Zhou","Pengyu Wang","Xian Li","Ying Fang","Yujie Yang","Xiaofei Li"],"abstract":"In this work, we propose CleanMel, a single-channel Mel-spectrogram denoising and dereverberation network for improving both speech quality and automatic speech recognition (ASR) performance. The proposed network takes as input the noisy and reverberant microphone recording and predicts the corresponding clean Mel-spectrogram. The enhanced Mel-spectrogram can be either transformed to speech waveform with a neural vocoder or directly used for ASR. The proposed network is composed of interleaved cross-band and narrow-band processing in the Mel-frequency domain, for learning the full-band spectral pattern and the narrow-band properties of signals, respectively. Compared to linear-frequency domain or time-domain speech enhancement, the key advantage of Mel-spectrogram enhancement is that Mel-frequency presents speech in a more compact way and thus is easier to learn, which will benefit both speech quality and ASR. Experimental results on four English and one Chinese datasets demonstrate a significant improvement in both speech quality and ASR performance achieved by the proposed model. Code and audio examples of our model are available online in https://audio.westlake.edu.cn/Research/CleanMel.html.","url_abs":"https://arxiv.org/abs/2502.20040v1","url_pdf":"https://arxiv.org/pdf/2502.20040v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"cleanmel-mel-spectrogram-enhancement-for","repo_url":"https://github.com/Audio-WestlakeU/CleanMel","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"automatic-speech-recognition-2","task_name":"Automatic Speech Recognition"},{"task_slug":"automatic-speech-recognition","task_name":"Automatic Speech Recognition (ASR)"},{"task_slug":"denoising","task_name":"Denoising"},{"task_slug":"speech-enhancement","task_name":"Speech Enhancement"},{"task_slug":"speech-recognition","task_name":"Speech Recognition"},{"task_slug":"speech-recognition-1","task_name":"speech-recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/automatic-speech-recognition-asr-on-realman","task":"Automatic Speech Recognition (ASR)","dataset":"RealMAN","model":"CleanMel-L-mask","rank_in_archive_order":1,"of":2,"metrics":{"CER":"14.4"},"uses_additional_data":false},{"leaderboard":"/sota/speech-enhancement-on-realman","task":"Speech Enhancement","dataset":"RealMAN","model":"CleanMel-L-map","rank_in_archive_order":1,"of":2,"metrics":{"DNSMOS":"3.82","DNSMOS BAK":"4.03","DNSMOS OVRL":"3.25","DNSMOS SIG":"3.55","PESQ-WB":"2.10"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}