{"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/style-tokens-unsupervised-style-modeling","title":"Style Tokens: Unsupervised Style Modeling, Control and Transfer in End-to-End Speech Synthesis","arxiv_id":"1803.09017","date":"2018-03-23","proceeding":"ICML 2018 7","authors":["Yuxuan Wang","Daisy Stanton","Yu Zhang","RJ Skerry-Ryan","Eric Battenberg","Joel Shor","Ying Xiao","Fei Ren","Ye Jia","Rif A. Saurous"],"abstract":"In this work, we propose \"global style tokens\" (GSTs), a bank of embeddings\nthat are jointly trained within Tacotron, a state-of-the-art end-to-end speech\nsynthesis system. The embeddings are trained with no explicit labels, yet learn\nto model a large range of acoustic expressiveness. GSTs lead to a rich set of\nsignificant results. The soft interpretable \"labels\" they generate can be used\nto control synthesis in novel ways, such as varying speed and speaking style -\nindependently of the text content. They can also be used for style transfer,\nreplicating the speaking style of a single audio clip across an entire\nlong-form text corpus. When trained on noisy, unlabeled found data, GSTs learn\nto factorize noise and speaker identity, providing a path towards highly\nscalable but robust speech synthesis.","url_abs":"http://arxiv.org/abs/1803.09017v1","url_pdf":"http://arxiv.org/pdf/1803.09017v1.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":"style-tokens-unsupervised-style-modeling","repo_url":"https://github.com/KinglittleQ/GST-Tacotron","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"style-tokens-unsupervised-style-modeling","repo_url":"https://github.com/Kyubyong/expressive_tacotron","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"style-tokens-unsupervised-style-modeling","repo_url":"https://github.com/acetylSv/GST-tacotron","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"style-tokens-unsupervised-style-modeling","repo_url":"https://github.com/cnlinxi/style-token_tacotron2","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"style-tokens-unsupervised-style-modeling","repo_url":"https://github.com/foamliu/GST-Tacotron-v2","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"style-tokens-unsupervised-style-modeling","repo_url":"https://github.com/hash2430/pitchtron","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"style-tokens-unsupervised-style-modeling","repo_url":"https://github.com/syang1993/gst-tacotron","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"style-tokens-unsupervised-style-modeling","repo_url":"https://github.com/CODEJIN/GST_Tacotron","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"style-tokens-unsupervised-style-modeling","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":"style-tokens-unsupervised-style-modeling","repo_url":"https://github.com/jinhan/tacotron2-gst","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"BSD-3-Clause"}},{"paper_slug":"style-tokens-unsupervised-style-modeling","repo_url":"https://github.com/keonlee9420/Cross-Speaker-Emotion-Transfer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"speech-synthesis","task_name":"Speech Synthesis"},{"task_slug":"style-transfer","task_name":"Style Transfer"},{"task_slug":"text-to-speech-synthesis","task_name":"Text-To-Speech Synthesis"}],"methods":[{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"bigru","method_name":"BiGRU"},{"method_slug":"cbhg","method_name":"CBHG"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"gru","method_name":"GRU"},{"method_slug":"griffin-lim-algorithm","method_name":"Griffin-Lim Algorithm"},{"method_slug":"highway-layer","method_name":"Highway Layer"},{"method_slug":"highway-network","method_name":"Highway Network"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"residual-gru","method_name":"Residual GRU"},{"method_slug":"speed","method_name":"SPEED"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tacotron","method_name":"Tacotron"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1803.09017","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1803.09017"}},"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/cnlinxi/style-token_tacotron2","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/acetylSv/GST-tacotron","reach":{"status":"ok"}},{"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/keonlee9420/Cross-Speaker-Emotion-Transfer","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/syang1993/gst-tacotron","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/foamliu/GST-Tacotron-v2","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/jinhan/tacotron2-gst","reach":{"status":"ok","spdx":"BSD-3-Clause"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/hash2430/pitchtron","reach":{"status":"ok","spdx":"NOASSERTION"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Kyubyong/expressive_tacotron","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/KinglittleQ/GST-Tacotron","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/CODEJIN/GST_Tacotron","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":4,"ran_fixture":1,"ran":1,"unverified":15},"by_repo_kind":{"listed":{"samples":20,"ran":5,"repositories":6}},"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":7,"samples":[{"code_sha256_prefix":"83e2c862f7374ac7","entry":"dynamic_range_compression","repo":"foamliu/GST-Tacotron-v2","repo_kind":"listed","path":"audio_processing.py","file_url":"https://github.com/foamliu/GST-Tacotron-v2/blob/HEAD/audio_processing.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"83e2c862f7374ac7"}},{"code_sha256_prefix":"b38f9c28c5117398","entry":"fused_add_tanh_sigmoid_multiply","repo":"foamliu/GST-Tacotron-v2","repo_kind":"listed","path":"glow.py","file_url":"https://github.com/foamliu/GST-Tacotron-v2/blob/HEAD/glow.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b38f9c28c5117398"}},{"code_sha256_prefix":"9f9ec26d4cdfdf7d","entry":"griffin_lim","repo":"jinhan/tacotron2-gst","repo_kind":"listed","path":"audio_processing.py","file_url":"https://github.com/jinhan/tacotron2-gst/blob/HEAD/audio_processing.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":false,"mcp_get_code":{"code_sha256":"9f9ec26d4cdfdf7d"}},{"code_sha256_prefix":"3494c70cf20413e6","entry":"looper","repo":"Kyubyong/expressive_tacotron","repo_kind":"listed","path":"synthesize.py","file_url":"https://github.com/Kyubyong/expressive_tacotron/blob/HEAD/synthesize.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"3494c70cf20413e6"}},{"code_sha256_prefix":"b13a72233912c0b1","entry":"remove","repo":"foamliu/GST-Tacotron-v2","repo_kind":"listed","path":"glow.py","file_url":"https://github.com/foamliu/GST-Tacotron-v2/blob/HEAD/glow.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":"b13a72233912c0b1"}},{"code_sha256_prefix":"b1491d3c2e709fdc","entry":"split_func","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"b1491d3c2e709fdc"}},{"code_sha256_prefix":"6611d63ef3343fb1","entry":"Get_Path","repo":"CODEJIN/GST_Tacotron","repo_kind":"listed","path":"Get_Path.py","file_url":"https://github.com/CODEJIN/GST_Tacotron/blob/HEAD/Get_Path.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":"6611d63ef3343fb1"}},{"code_sha256_prefix":"8b84f60b5ff9e153","entry":"apply_gradient_allreduce","repo":"jinhan/tacotron2-gst","repo_kind":"listed","path":"distributed.py","file_url":"https://github.com/jinhan/tacotron2-gst/blob/HEAD/distributed.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":false,"mcp_get_code":{"code_sha256":"8b84f60b5ff9e153"}},{"code_sha256_prefix":"f887d5ffcbc6ad3b","entry":"conversion_helper","repo":"jinhan/tacotron2-gst","repo_kind":"listed","path":"fp16_optimizer.py","file_url":"https://github.com/jinhan/tacotron2-gst/blob/HEAD/fp16_optimizer.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":false,"mcp_get_code":{"code_sha256":"f887d5ffcbc6ad3b"}},{"code_sha256_prefix":"38034aa1d784464d","entry":"fp16_to_fp32","repo":"jinhan/tacotron2-gst","repo_kind":"listed","path":"fp16_optimizer.py","file_url":"https://github.com/jinhan/tacotron2-gst/blob/HEAD/fp16_optimizer.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":false,"mcp_get_code":{"code_sha256":"38034aa1d784464d"}},{"code_sha256_prefix":"ea22baceb9a219a1","entry":"fp32_to_fp16","repo":"jinhan/tacotron2-gst","repo_kind":"listed","path":"fp16_optimizer.py","file_url":"https://github.com/jinhan/tacotron2-gst/blob/HEAD/fp16_optimizer.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":false,"mcp_get_code":{"code_sha256":"ea22baceb9a219a1"}},{"code_sha256_prefix":"e274cfa39f3d6bf9","entry":"griffin_lim","repo":"foamliu/GST-Tacotron-v2","repo_kind":"listed","path":"audio_processing.py","file_url":"https://github.com/foamliu/GST-Tacotron-v2/blob/HEAD/audio_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":"e274cfa39f3d6bf9"}},{"code_sha256_prefix":"bd89c559f716b290","entry":"inv_preemphasis","repo":"cnlinxi/style-token_tacotron2","repo_kind":"listed","path":"datasets/audio.py","file_url":"https://github.com/cnlinxi/style-token_tacotron2/blob/HEAD/datasets/audio.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":"bd89c559f716b290"}},{"code_sha256_prefix":"f1b4c5cfbaef7ff5","entry":"inv_preemphasis","repo":"CODEJIN/GST_Tacotron","repo_kind":"listed","path":"Audio.py","file_url":"https://github.com/CODEJIN/GST_Tacotron/blob/HEAD/Audio.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":"f1b4c5cfbaef7ff5"}},{"code_sha256_prefix":"203a81d997da01e5","entry":"is_mulaw","repo":"cnlinxi/style-token_tacotron2","repo_kind":"listed","path":"wavenet_vocoder/util.py","file_url":"https://github.com/cnlinxi/style-token_tacotron2/blob/HEAD/wavenet_vocoder/util.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":"203a81d997da01e5"}},{"code_sha256_prefix":"9f7f281fd2dc3db3","entry":"is_mulaw_quantize","repo":"cnlinxi/style-token_tacotron2","repo_kind":"listed","path":"wavenet_vocoder/util.py","file_url":"https://github.com/cnlinxi/style-token_tacotron2/blob/HEAD/wavenet_vocoder/util.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":"9f7f281fd2dc3db3"}},{"code_sha256_prefix":"0ce2fef6ac56c0bd","entry":"is_raw","repo":"cnlinxi/style-token_tacotron2","repo_kind":"listed","path":"wavenet_vocoder/util.py","file_url":"https://github.com/cnlinxi/style-token_tacotron2/blob/HEAD/wavenet_vocoder/util.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":"0ce2fef6ac56c0bd"}},{"code_sha256_prefix":"d3e8204b867d34ac","entry":"load_model","repo":"KinglittleQ/GST-Tacotron","repo_kind":"listed","path":"generate.py","file_url":"https://github.com/KinglittleQ/GST-Tacotron/blob/HEAD/generate.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":"d3e8204b867d34ac"}},{"code_sha256_prefix":"ae38e046c059c78f","entry":"norm_data","repo":"cnlinxi/style-token_tacotron2","repo_kind":"listed","path":"preprocess.py","file_url":"https://github.com/cnlinxi/style-token_tacotron2/blob/HEAD/preprocess.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":"ae38e046c059c78f"}},{"code_sha256_prefix":"56a333de6e89087a","entry":"preemphasis","repo":"cnlinxi/style-token_tacotron2","repo_kind":"listed","path":"datasets/audio.py","file_url":"https://github.com/cnlinxi/style-token_tacotron2/blob/HEAD/datasets/audio.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":"56a333de6e89087a"}},{"code_sha256_prefix":"31d5837f91baf6c1","entry":"preemphasis","repo":"CODEJIN/GST_Tacotron","repo_kind":"listed","path":"Audio.py","file_url":"https://github.com/CODEJIN/GST_Tacotron/blob/HEAD/Audio.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":"31d5837f91baf6c1"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}