{"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/panns-large-scale-pretrained-audio-neural-1","title":"PANNs: Large-Scale Pretrained Audio Neural Networks for Audio Pattern Recognition","arxiv_id":"1912.10211","date":"2019-12-21","proceeding":null,"authors":["Qiuqiang Kong","Yin Cao","Turab Iqbal","Yuxuan Wang","Wenwu Wang","Mark D. Plumbley"],"abstract":"Audio pattern recognition is an important research topic in the machine learning area, and includes several tasks such as audio tagging, acoustic scene classification, music classification, speech emotion classification and sound event detection. Recently, neural networks have been applied to tackle audio pattern recognition problems. However, previous systems are built on specific datasets with limited durations. Recently, in computer vision and natural language processing, systems pretrained on large-scale datasets have generalized well to several tasks. However, there is limited research on pretraining systems on large-scale datasets for audio pattern recognition. In this paper, we propose pretrained audio neural networks (PANNs) trained on the large-scale AudioSet dataset. These PANNs are transferred to other audio related tasks. We investigate the performance and computational complexity of PANNs modeled by a variety of convolutional neural networks. We propose an architecture called Wavegram-Logmel-CNN using both log-mel spectrogram and waveform as input feature. Our best PANN system achieves a state-of-the-art mean average precision (mAP) of 0.439 on AudioSet tagging, outperforming the best previous system of 0.392. We transfer PANNs to six audio pattern recognition tasks, and demonstrate state-of-the-art performance in several of those tasks. We have released the source code and pretrained models of PANNs: https://github.com/qiuqiangkong/audioset_tagging_cnn.","url_abs":"http://arxiv.org/abs/1912.10211v5","url_pdf":"http://arxiv.org/pdf/1912.10211v5.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":"links_only","authors_date_abstract":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license), from the Kaggle arXiv metadata snapshot of 2026-09-12"},"code_links":[{"paper_slug":"panns-large-scale-pretrained-audio-neural-1","repo_url":"https://github.com/yinkalario/General-Purpose-Sound-Recognition-Demo","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"GPL-3.0"}},{"paper_slug":"panns-large-scale-pretrained-audio-neural-1","repo_url":"https://github.com/qiuqiangkong/audioset_tagging_cnn","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"panns-large-scale-pretrained-audio-neural-1","repo_url":"https://github.com/YuanGongND/psla","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"panns-large-scale-pretrained-audio-neural-1","repo_url":"https://github.com/google-research/leaf-audio","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"panns-large-scale-pretrained-audio-neural-1","repo_url":"https://github.com/gudgud96/frechet-audio-distance","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"panns-large-scale-pretrained-audio-neural-1","repo_url":"https://github.com/nttcslab/composing-general-audio-repr","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"panns-large-scale-pretrained-audio-neural-1","repo_url":"https://github.com/PaddlePaddle/PaddleSpeech","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"paddle","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"panns-large-scale-pretrained-audio-neural-1","repo_url":"https://github.com/sithu31296/audio-tagging","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/audio-classification-on-audioset","task":"Audio Classification","dataset":"AudioSet","model":"PANNs-CNN14 (Single)","rank_in_archive_order":41,"of":51,"metrics":{"AUC":"0.973","Test mAP":"0.431","d-prime":"2.732"},"uses_additional_data":false},{"leaderboard":"/sota/audio-tagging-on-audioset","task":"Audio Tagging","dataset":"AudioSet","model":"CNN14","rank_in_archive_order":11,"of":11,"metrics":{"mean average precision":"0.431"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1912.10211","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1912.10211"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/qiuqiangkong/audioset_tagging_cnn","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/PaddlePaddle/PaddleSpeech","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/gudgud96/frechet-audio-distance","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/yinkalario/General-Purpose-Sound-Recognition-Demo","reach":{"status":"ok","spdx":"GPL-3.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/nttcslab/composing-general-audio-repr","reach":{"status":"ok","spdx":"NOASSERTION"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/google-research/leaf-audio","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/YuanGongND/psla","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/sithu31296/audio-tagging","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":2,"unverified":14},"by_repo_kind":{"official":{"samples":7,"ran":2,"repositories":1},"listed":{"samples":9,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"b488d4a95f800c00","entry":"do_mixup","repo":"qiuqiangkong/audioset_tagging_cnn","repo_kind":"official","path":"pytorch/pytorch_utils.py","file_url":"https://github.com/qiuqiangkong/audioset_tagging_cnn/blob/HEAD/pytorch/pytorch_utils.py","link_basis":"plan_row","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b488d4a95f800c00"}},{"code_sha256_prefix":"6e703419174005ee","entry":"move_data_to_device","repo":"qiuqiangkong/audioset_tagging_cnn","repo_kind":"official","path":"pytorch/pytorch_utils.py","file_url":"https://github.com/qiuqiangkong/audioset_tagging_cnn/blob/HEAD/pytorch/pytorch_utils.py","link_basis":"plan_row","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6e703419174005ee"}},{"code_sha256_prefix":"d051ddaa6aa3a0ee","entry":"clip_bce","repo":"qiuqiangkong/audioset_tagging_cnn","repo_kind":"official","path":"pytorch/losses.py","file_url":"https://github.com/qiuqiangkong/audioset_tagging_cnn/blob/HEAD/pytorch/losses.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d051ddaa6aa3a0ee"}},{"code_sha256_prefix":"3e405654b0944236","entry":"collate_fn","repo":"qiuqiangkong/audioset_tagging_cnn","repo_kind":"official","path":"utils/data_generator.py","file_url":"https://github.com/qiuqiangkong/audioset_tagging_cnn/blob/HEAD/utils/data_generator.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"3e405654b0944236"}},{"code_sha256_prefix":"2376b34dc0f7c346","entry":"count_parameters","repo":"sithu31296/audio-tagging","repo_kind":"listed","path":"utils/utils.py","file_url":"https://github.com/sithu31296/audio-tagging/blob/HEAD/utils/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"2376b34dc0f7c346"}},{"code_sha256_prefix":"3a101004e6f0c3e0","entry":"forward","repo":"qiuqiangkong/audioset_tagging_cnn","repo_kind":"official","path":"pytorch/pytorch_utils.py","file_url":"https://github.com/qiuqiangkong/audioset_tagging_cnn/blob/HEAD/pytorch/pytorch_utils.py","link_basis":"plan_row","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"3a101004e6f0c3e0"}},{"code_sha256_prefix":"a70ac1a79860cc57","entry":"get_loss","repo":"sithu31296/audio-tagging","repo_kind":"listed","path":"utils/losses.py","file_url":"https://github.com/sithu31296/audio-tagging/blob/HEAD/utils/losses.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a70ac1a79860cc57"}},{"code_sha256_prefix":"a79c3eb46da2b363","entry":"get_loss_func","repo":"qiuqiangkong/audioset_tagging_cnn","repo_kind":"official","path":"pytorch/losses.py","file_url":"https://github.com/qiuqiangkong/audioset_tagging_cnn/blob/HEAD/pytorch/losses.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a79c3eb46da2b363"}},{"code_sha256_prefix":"93c4d29ae02a17c2","entry":"get_model_size","repo":"sithu31296/audio-tagging","repo_kind":"listed","path":"utils/utils.py","file_url":"https://github.com/sithu31296/audio-tagging/blob/HEAD/utils/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"93c4d29ae02a17c2"}},{"code_sha256_prefix":"e7ec50eb80597850","entry":"get_optimizer","repo":"sithu31296/audio-tagging","repo_kind":"listed","path":"utils/optimizers.py","file_url":"https://github.com/sithu31296/audio-tagging/blob/HEAD/utils/optimizers.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e7ec50eb80597850"}},{"code_sha256_prefix":"d966b9be6af68778","entry":"get_scheduler","repo":"sithu31296/audio-tagging","repo_kind":"listed","path":"utils/schedulers.py","file_url":"https://github.com/sithu31296/audio-tagging/blob/HEAD/utils/schedulers.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d966b9be6af68778"}},{"code_sha256_prefix":"5b197f92e336b6c7","entry":"interpolate","repo":"sithu31296/audio-tagging","repo_kind":"listed","path":"models/cnn14.py","file_url":"https://github.com/sithu31296/audio-tagging/blob/HEAD/models/cnn14.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5b197f92e336b6c7"}},{"code_sha256_prefix":"4510c24e528ea1bf","entry":"mixup_augment","repo":"sithu31296/audio-tagging","repo_kind":"listed","path":"datasets/transforms.py","file_url":"https://github.com/sithu31296/audio-tagging/blob/HEAD/datasets/transforms.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4510c24e528ea1bf"}},{"code_sha256_prefix":"91592112bbcf4758","entry":"pad_framewise","repo":"sithu31296/audio-tagging","repo_kind":"listed","path":"models/cnn14.py","file_url":"https://github.com/sithu31296/audio-tagging/blob/HEAD/models/cnn14.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"91592112bbcf4758"}},{"code_sha256_prefix":"175ae0516204fc91","entry":"read_black_list","repo":"qiuqiangkong/audioset_tagging_cnn","repo_kind":"official","path":"utils/data_generator.py","file_url":"https://github.com/qiuqiangkong/audioset_tagging_cnn/blob/HEAD/utils/data_generator.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"175ae0516204fc91"}},{"code_sha256_prefix":"9251df088bafc792","entry":"test_model_latency","repo":"sithu31296/audio-tagging","repo_kind":"listed","path":"utils/utils.py","file_url":"https://github.com/sithu31296/audio-tagging/blob/HEAD/utils/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9251df088bafc792"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}