{"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/bigvgan-a-universal-neural-vocoder-with-large","title":"BigVGAN: A Universal Neural Vocoder with Large-Scale Training","arxiv_id":"2206.04658","date":"2022-06-09","proceeding":null,"authors":["Sang-gil Lee","Wei Ping","Boris Ginsburg","Bryan Catanzaro","Sungroh Yoon"],"abstract":"Despite recent progress in generative adversarial network (GAN)-based vocoders, where the model generates raw waveform conditioned on acoustic features, it is challenging to synthesize high-fidelity audio for numerous speakers across various recording environments. In this work, we present BigVGAN, a universal vocoder that generalizes well for various out-of-distribution scenarios without fine-tuning. We introduce periodic activation function and anti-aliased representation into the GAN generator, which brings the desired inductive bias for audio synthesis and significantly improves audio quality. In addition, we train our GAN vocoder at the largest scale up to 112M parameters, which is unprecedented in the literature. We identify and address the failure modes in large-scale GAN training for audio, while maintaining high-fidelity output without over-regularization. Our BigVGAN, trained only on clean speech (LibriTTS), achieves the state-of-the-art performance for various zero-shot (out-of-distribution) conditions, including unseen speakers, languages, recording environments, singing voices, music, and instrumental audio. We release our code and model at: https://github.com/NVIDIA/BigVGAN","url_abs":"https://arxiv.org/abs/2206.04658v2","url_pdf":"https://arxiv.org/pdf/2206.04658v2.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":"bigvgan-a-universal-neural-vocoder-with-large","repo_url":"https://github.com/nvidia/bigvgan","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"bigvgan-a-universal-neural-vocoder-with-large","repo_url":"https://github.com/sh-lee-prml/BigVGAN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"bigvgan-a-universal-neural-vocoder-with-large","repo_url":"https://github.com/sh-lee-prml/hierspeechpp","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"bigvgan-a-universal-neural-vocoder-with-large","repo_url":"https://github.com/sh-lee-prml/periodwave","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"bigvgan-a-universal-neural-vocoder-with-large","repo_url":"https://github.com/IAHispano/Applio/tree/exp/vocoders/rvc/lib/algorithm","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"audio-generation","task_name":"Audio Generation"},{"task_slug":"audio-synthesis","task_name":"Audio Synthesis"},{"task_slug":null,"task_name":"Generative Adversarial Network"},{"task_slug":"inductive-bias","task_name":"Inductive Bias"},{"task_slug":"music-generation","task_name":"Music Generation"},{"task_slug":"speech-synthesis","task_name":"Speech Synthesis"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/speech-synthesis-on-libritts","task":"Speech Synthesis","dataset":"LibriTTS","model":"BigVGAN-v2","rank_in_archive_order":2,"of":15,"metrics":{"M-STFT":"0.7026","MCD":"0.2903","PESQ":"4.362","Periodicity":"0.0593","V/UV F1":"0.9793"},"uses_additional_data":true},{"leaderboard":"/sota/speech-synthesis-on-libritts","task":"Speech Synthesis","dataset":"LibriTTS","model":"BigVGAN","rank_in_archive_order":9,"of":15,"metrics":{"M-STFT":"0.7997","MCD":"0.3745","PESQ":"4.027","Periodicity":"0.1018","V/UV F1":"0.9598"},"uses_additional_data":false},{"leaderboard":"/sota/speech-synthesis-on-libritts","task":"Speech Synthesis","dataset":"LibriTTS","model":"BigVGAN-base","rank_in_archive_order":11,"of":15,"metrics":{"M-STFT":"0.8788","MCD":"0.4564","PESQ":"3.519","Periodicity":"0.1287","V/UV F1":"0.9459"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2206.04658","atlas_url":"https://app.syntology.ai/?focus=2206.04658","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2206.04658"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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":"deterministic:regex_extraction","url":"https://github.com/NVIDIA/BigVGAN","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/nvidia/bigvgan","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/IAHispano/Applio/tree/exp/vocoders/rvc/lib/algorithm","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/sh-lee-prml/BigVGAN","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/sh-lee-prml/periodwave","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/sh-lee-prml/hierspeechpp","reach":null}],"summary":{"ran":4,"ran_draft_wrong":1,"ran_honours":3,"unverified":9},"by_repo_kind":{"listed":{"samples":16,"ran":7,"repositories":3},"community":{"samples":1,"ran":1,"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":1,"samples":[{"code_sha256_prefix":"90e3f68179747adc","entry":"DiscriminatorP","repo":"sh-lee-prml/periodwave","repo_kind":"listed","path":"model/bigvganv2_discriminator.py","file_url":"https://github.com/sh-lee-prml/periodwave/blob/HEAD/model/bigvganv2_discriminator.py","link_basis":"first_harvest_node","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":"90e3f68179747adc"}},{"code_sha256_prefix":"655dde631187450a","entry":"MultiPeriodDiscriminator","repo":"sh-lee-prml/periodwave","repo_kind":"listed","path":"model/bigvganv2_discriminator.py","file_url":"https://github.com/sh-lee-prml/periodwave/blob/HEAD/model/bigvganv2_discriminator.py","link_basis":"first_harvest_node","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":"655dde631187450a"}},{"code_sha256_prefix":"02d0dcf94b91a8ac","entry":"ResBlock1","repo":"sh-lee-prml/hierspeechpp","repo_kind":"listed","path":"modules.py","file_url":"https://github.com/sh-lee-prml/hierspeechpp/blob/HEAD/modules.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"02d0dcf94b91a8ac"}},{"code_sha256_prefix":"dabc5c8e7c1ae006","entry":"clones","repo":"sh-lee-prml/BigVGAN","repo_kind":"listed","path":"model_vits_with_bigvgan.py","file_url":"https://github.com/sh-lee-prml/BigVGAN/blob/HEAD/model_vits_with_bigvgan.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"dabc5c8e7c1ae006"}},{"code_sha256_prefix":"59ae0708b1e8d94b","entry":"convert_pad_shape","repo":"sh-lee-prml/BigVGAN","repo_kind":"listed","path":"commons.py","file_url":"https://github.com/sh-lee-prml/BigVGAN/blob/HEAD/commons.py","link_basis":"harvester_set","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"59ae0708b1e8d94b"}},{"code_sha256_prefix":"a26f85d7c72ef39a","entry":"get_padding","repo":"sh-lee-prml/BigVGAN","repo_kind":"listed","path":"commons.py","file_url":"https://github.com/sh-lee-prml/BigVGAN/blob/HEAD/commons.py","link_basis":"harvester_set","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a26f85d7c72ef39a"}},{"code_sha256_prefix":"f1cba67eb6d6e4ad","entry":"get_param_num","repo":"Zz-ww/VITS-BigVGAN-SpanPSP-Chinese","repo_kind":"community","path":"train_vits_with_bigvgan.py","file_url":"https://github.com/Zz-ww/VITS-BigVGAN-SpanPSP-Chinese/blob/HEAD/train_vits_with_bigvgan.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f1cba67eb6d6e4ad"}},{"code_sha256_prefix":"70a7614b3b3eb41f","entry":"unconstrained_rational_quadratic_spline","repo":"sh-lee-prml/BigVGAN","repo_kind":"listed","path":"transforms.py","file_url":"https://github.com/sh-lee-prml/BigVGAN/blob/HEAD/transforms.py","link_basis":"harvester_set","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":"70a7614b3b3eb41f"}},{"code_sha256_prefix":"896f7aae82a62114","entry":"discriminator_loss","repo":"sh-lee-prml/BigVGAN","repo_kind":"listed","path":"losses.py","file_url":"https://github.com/sh-lee-prml/BigVGAN/blob/HEAD/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":"896f7aae82a62114"}},{"code_sha256_prefix":"262807b9eb9706a5","entry":"dynamic_range_compression_torch","repo":"sh-lee-prml/BigVGAN","repo_kind":"listed","path":"mel_processing.py","file_url":"https://github.com/sh-lee-prml/BigVGAN/blob/HEAD/mel_processing.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":"262807b9eb9706a5"}},{"code_sha256_prefix":"78c12e563d5af53d","entry":"dynamic_range_decompression_torch","repo":"sh-lee-prml/BigVGAN","repo_kind":"listed","path":"mel_processing.py","file_url":"https://github.com/sh-lee-prml/BigVGAN/blob/HEAD/mel_processing.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":"78c12e563d5af53d"}},{"code_sha256_prefix":"24e8b4731dc3e070","entry":"feature_loss","repo":"sh-lee-prml/BigVGAN","repo_kind":"listed","path":"losses.py","file_url":"https://github.com/sh-lee-prml/BigVGAN/blob/HEAD/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":"24e8b4731dc3e070"}},{"code_sha256_prefix":"6191fd9efbdd55d4","entry":"generator_loss","repo":"sh-lee-prml/BigVGAN","repo_kind":"listed","path":"losses.py","file_url":"https://github.com/sh-lee-prml/BigVGAN/blob/HEAD/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":"6191fd9efbdd55d4"}},{"code_sha256_prefix":"c01a3efb274ff778","entry":"intersperse","repo":"sh-lee-prml/BigVGAN","repo_kind":"listed","path":"commons.py","file_url":"https://github.com/sh-lee-prml/BigVGAN/blob/HEAD/commons.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":"c01a3efb274ff778"}},{"code_sha256_prefix":"ba6cafc4d658c14f","entry":"piecewise_rational_quadratic_transform","repo":"sh-lee-prml/BigVGAN","repo_kind":"listed","path":"transforms.py","file_url":"https://github.com/sh-lee-prml/BigVGAN/blob/HEAD/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":"ba6cafc4d658c14f"}},{"code_sha256_prefix":"e85d747983c4f5d8","entry":"searchsorted","repo":"sh-lee-prml/BigVGAN","repo_kind":"listed","path":"transforms.py","file_url":"https://github.com/sh-lee-prml/BigVGAN/blob/HEAD/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":"e85d747983c4f5d8"}},{"code_sha256_prefix":"26b87488d429896a","entry":"spectral_normalize_torch","repo":"sh-lee-prml/BigVGAN","repo_kind":"listed","path":"mel_processing.py","file_url":"https://github.com/sh-lee-prml/BigVGAN/blob/HEAD/mel_processing.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":"26b87488d429896a"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}