{"url":"/task/speech-to-speech-translation","name":"Speech-to-Speech Translation","slug":"speech-to-speech-translation","description_markdown":"Speech-to-speech translation (S2ST) consists on translating speech from one language to speech in another language. This can be done with a cascade of automatic speech recognition (ASR), text-to-text machine translation (MT), and text-to-speech (TTS) synthesis\r\nsub-systems, which is text-centric. Recently, works on S2ST without relying on intermediate text representation is emerging.","categories":[{"name":"Speech","url":"/area/speech"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":117,"papers_with_code":39,"benchmarks":3,"benchmark_tables_in_archive":3,"benchmark_tables_shown":3,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":5,"subtasks":0,"parent_tasks":0},"benchmarks":[{"leaderboard":"/sota/speech-to-speech-translation-on-tat","slug":"speech-to-speech-translation-on-tat","dataset":"TAT","dataset_url":"/dataset/tat","rows_in_archive":8,"metrics":["ASR-BLEU (Dev)","ASR-BLEU (Test)"],"first_row_in_archive_order":{"model":"Hokkien→En (Two-pass decoding)","paper_title":"Speech-to-speech translation for a real-world unwritten language","paper_url":"/paper/speech-to-speech-translation-for-a-real-world","paper_date":"2022-10-19","arxiv_id":null,"code_links":[{"title":"facebookresearch/fairseq","url":"https://github.com/facebookresearch/fairseq/tree/ust/examples/hokkien"}],"syntology":null}},{"leaderboard":"/sota/speech-to-speech-translation-on-fleurs-x-eng","slug":"speech-to-speech-translation-on-fleurs-x-eng","dataset":"FLEURS X-eng","dataset_url":null,"rows_in_archive":7,"metrics":["ASR-BLEU"],"first_row_in_archive_order":{"model":"GenTranslateV2","paper_title":"GenTranslate: Large Language Models are Generative Multilingual Speech and Machine Translators","paper_url":"/paper/gentranslate-large-language-models-are","paper_date":"2024-02-10","arxiv_id":"2402.06894","code_links":[{"title":"yuchen005/gentranslate","url":"https://github.com/yuchen005/gentranslate"}],"syntology":null}},{"leaderboard":"/sota/speech-to-speech-translation-on-cvss","slug":"speech-to-speech-translation-on-cvss","dataset":"CVSS","dataset_url":"/dataset/cvss","rows_in_archive":2,"metrics":["ASR-BLEU","Parameters"],"first_row_in_archive_order":{"model":"SeamlessM4T Large","paper_title":"SeamlessM4T: Massively Multilingual & Multimodal Machine Translation","paper_url":"/paper/seamlessm4t-massively-multilingual-multimodal","paper_date":"2023-08-22","arxiv_id":"2308.11596","code_links":[{"title":"facebookresearch/seamless_communication","url":"https://github.com/facebookresearch/seamless_communication"},{"title":"facebookresearch/sonar","url":"https://github.com/facebookresearch/sonar"},{"title":"yangyucheng000/University","url":"https://github.com/yangyucheng000/University/tree/main/model-3/seamless_m4t"},{"title":"pwc-1/Paper-10","url":"https://github.com/pwc-1/Paper-10/tree/main/seamless_m4t"}],"syntology":{"n":2,"n_ran":2,"n_unverified":0,"n_pointer_only":2}}}],"datasets":[{"url":"/dataset/cvss","name":"CVSS","full_name":"","num_papers_in_archive":26},{"url":"/dataset/speechmatrix","name":"SpeechMatrix","full_name":"","num_papers_in_archive":6},{"url":"/dataset/heroes-corpus","name":"Heroes Corpus","full_name":"","num_papers_in_archive":1},{"url":"/dataset/libris2s","name":"LibriS2S","full_name":"","num_papers_in_archive":1},{"url":"/dataset/tat","name":"TAT","full_name":"Taiwanese Across Taiwan","num_papers_in_archive":1}],"subtasks":[],"parent_tasks":[],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":30,"of":39,"tagged_in_all":117,"items":[{"url":"/paper/robust-speech-recognition-via-large-scale-1","title":"Robust Speech Recognition via Large-Scale Weak Supervision","date":"2022-12-06","arxiv_id":"2212.04356","repositories_listed":15,"syntology":{"n":59,"n_ran":5,"n_unverified":54,"n_pointer_only":18}},{"url":"/paper/audiolm-a-language-modeling-approach-to-audio","title":"AudioLM: a Language Modeling Approach to Audio Generation","date":"2022-09-07","arxiv_id":"2209.03143","repositories_listed":6,"syntology":{"n":15,"n_ran":4,"n_unverified":11,"n_pointer_only":3}},{"url":"/paper/seamlessm4t-massively-multilingual-multimodal","title":"SeamlessM4T: Massively Multilingual & Multimodal Machine Translation","date":"2023-08-22","arxiv_id":"2308.11596","repositories_listed":4,"syntology":{"n":2,"n_ran":2,"n_unverified":0,"n_pointer_only":2}},{"url":"/paper/funaudiollm-voice-understanding-and","title":"FunAudioLLM: Voice Understanding and Generation Foundation Models for Natural Interaction Between Humans and LLMs","date":"2024-07-04","arxiv_id":"2407.04051","repositories_listed":3,"syntology":{"n":10,"n_ran":10,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/unit-based-speech-to-speech-translation","title":"Textless Speech-to-Speech Translation With Limited Parallel Data","date":"2023-05-24","arxiv_id":"2305.15405","repositories_listed":2,"syntology":null},{"url":"/paper/leveraging-unit-language-guidance-to-advance","title":"Leveraging Unit Language Guidance to Advance Speech Modeling in Textless Speech-to-Speech Translation","date":"2025-05-21","arxiv_id":"2505.15333","repositories_listed":1,"syntology":null},{"url":"/paper/using-phonemes-in-cascaded-s2s-translation","title":"Using Phonemes in cascaded S2S translation pipeline","date":"2025-04-22","arxiv_id":"2504.16234","repositories_listed":1,"syntology":null},{"url":"/paper/high-fidelity-simultaneous-speech-to-speech","title":"High-Fidelity Simultaneous Speech-To-Speech Translation","date":"2025-02-05","arxiv_id":"2502.03382","repositories_listed":1,"syntology":null},{"url":"/paper/improving-speech-emotion-recognition-in-under","title":"Improving Speech Emotion Recognition in Under-Resourced Languages via Speech-to-Speech Translation with Bootstrapping Data Selection","date":"2024-09-17","arxiv_id":"2409.10985","repositories_listed":1,"syntology":null},{"url":"/paper/laugh-now-cry-later-controlling-time-varying","title":"Laugh Now Cry Later: Controlling Time-Varying Emotional States of Flow-Matching-Based Zero-Shot Text-to-Speech","date":"2024-07-17","arxiv_id":"2407.12229","repositories_listed":1,"syntology":null},{"url":"/paper/ctc-based-non-autoregressive-textless-speech","title":"CTC-based Non-autoregressive Textless Speech-to-Speech Translation","date":"2024-06-11","arxiv_id":"2406.07330","repositories_listed":1,"syntology":null},{"url":"/paper/a-non-autoregressive-generation-framework-for","title":"A Non-autoregressive Generation Framework for End-to-End Simultaneous Speech-to-Speech Translation","date":"2024-06-11","arxiv_id":"2406.06937","repositories_listed":1,"syntology":{"n":2,"n_ran":2,"n_unverified":0,"n_pointer_only":2}},{"url":"/paper/streamspeech-simultaneous-speech-to-speech","title":"StreamSpeech: Simultaneous Speech-to-Speech Translation with Multi-task Learning","date":"2024-06-05","arxiv_id":"2406.03049","repositories_listed":1,"syntology":{"n":4,"n_ran":4,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/transvip-speech-to-speech-translation-system","title":"TransVIP: Speech to Speech Translation System with Voice and Isochrony Preservation","date":"2024-05-28","arxiv_id":"2405.17809","repositories_listed":1,"syntology":null},{"url":"/paper/diffnorm-self-supervised-normalization-for","title":"DiffNorm: Self-Supervised Normalization for Non-autoregressive Speech-to-speech Translation","date":"2024-05-22","arxiv_id":"2405.13274","repositories_listed":1,"syntology":null},{"url":"/paper/gentranslate-large-language-models-are","title":"GenTranslate: Large Language Models are Generative Multilingual Speech and Machine Translators","date":"2024-02-10","arxiv_id":"2402.06894","repositories_listed":1,"syntology":null},{"url":"/paper/emphassess-a-prosodic-benchmark-on-assessing","title":"EmphAssess : a Prosodic Benchmark on Assessing Emphasis Transfer in Speech-to-Speech Models","date":"2023-12-21","arxiv_id":"2312.14069","repositories_listed":1,"syntology":null},{"url":"/paper/av2av-direct-audio-visual-speech-to-audio","title":"AV2AV: Direct Audio-Visual Speech to Audio-Visual Speech Translation with Unified Audio-Visual Speech Representation","date":"2023-12-05","arxiv_id":"2312.02512","repositories_listed":1,"syntology":{"n":7,"n_ran":4,"n_unverified":3,"n_pointer_only":0}},{"url":"/paper/daspeech-directed-acyclic-transformer-for-1","title":"DASpeech: Directed Acyclic Transformer for Fast and High-quality Speech-to-Speech Translation","date":"2023-10-11","arxiv_id":"2310.07403","repositories_listed":1,"syntology":{"n":7,"n_ran":5,"n_unverified":2,"n_pointer_only":7}},{"url":"/paper/many-to-many-spoken-language-translation-via","title":"Textless Unit-to-Unit training for Many-to-Many Multilingual Speech-to-Speech Translation","date":"2023-08-03","arxiv_id":"2308.01831","repositories_listed":1,"syntology":null},{"url":"/paper/towards-cross-language-prosody-transfer-for","title":"Towards cross-language prosody transfer for dialog","date":"2023-07-09","arxiv_id":"2307.04123","repositories_listed":1,"syntology":null},{"url":"/paper/learning-when-to-speak-latency-and-quality","title":"Learning When to Speak: Latency and Quality Trade-offs for Simultaneous Speech-to-Speech Translation with Offline Models","date":"2023-06-01","arxiv_id":"2306.01201","repositories_listed":1,"syntology":null},{"url":"/paper/espnet-st-v2-multipurpose-spoken-language","title":"ESPnet-ST-v2: Multipurpose Spoken Language Translation Toolkit","date":"2023-04-10","arxiv_id":"2304.04596","repositories_listed":1,"syntology":{"n":3,"n_ran":3,"n_unverified":0,"n_pointer_only":1}},{"url":"/paper/speak-foreign-languages-with-your-own-voice","title":"Speak Foreign Languages with Your Own Voice: Cross-Lingual Neural Codec Language Modeling","date":"2023-03-07","arxiv_id":"2303.03926","repositories_listed":1,"syntology":{"n":6,"n_ran":4,"n_unverified":2,"n_pointer_only":0}},{"url":"/paper/blaser-a-text-free-speech-to-speech","title":"BLASER: A Text-Free Speech-to-Speech Translation Evaluation Metric","date":"2022-12-16","arxiv_id":"2212.08486","repositories_listed":1,"syntology":null},{"url":"/paper/unity-two-pass-direct-speech-to-speech","title":"UnitY: Two-pass Direct Speech-to-speech Translation with Discrete Units","date":"2022-12-15","arxiv_id":"2212.08055","repositories_listed":1,"syntology":null},{"url":"/paper/dialogs-re-enacted-across-languages","title":"Dialogs Re-enacted Across Languages","date":"2022-11-18","arxiv_id":"2211.11584","repositories_listed":1,"syntology":null},{"url":"/paper/joint-pre-training-with-speech-and-bilingual","title":"Joint Pre-Training with Speech and Bilingual Text for Direct Speech to Speech Translation","date":"2022-10-31","arxiv_id":"2210.17027","repositories_listed":1,"syntology":null},{"url":"/paper/a-textless-metric-for-speech-to-speech","title":"A Textless Metric for Speech-to-Speech Comparison","date":"2022-10-21","arxiv_id":"2210.11835","repositories_listed":1,"syntology":null},{"url":"/paper/transpeech-speech-to-speech-translation-with","title":"TranSpeech: Speech-to-Speech Translation With Bilateral Perturbation","date":"2022-05-25","arxiv_id":"2205.12523","repositories_listed":1,"syntology":null}],"syntology_records":10,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}