{"id":"fish-audio-s2-pro","nom":"Fish Audio S2 Pro","editeur":"Fish Audio","pays":"États-Unis","poids_ouverts":true,"poids_selon":null,"licence":"Fish Audio Research License","licence_selon":null,"liens":{"page":"https://huggingface.co/fishaudio/s2-pro","essayer":"https://fish.audio"},"fiche_auto":true,"page":"https://quelleia.com/voix/modeles/fish-audio-s2-pro/","releve":"2026-10-19","en_bref":"26e sur 31 en synthèse vocale en français, dans un peloton de 4 modèles que la marge d'erreur ne permet pas de séparer ; 2e sur 16 en prononciation du français ; 35e sur 96 en synthèse vocale en anglais (Artificial Analysis), dans un peloton de 7 modèles que la marge d'erreur ne permet pas de séparer.","presentation":"Modèle de synthèse vocale de Fish Audio à architecture double autorégressive, piloté par des indications libres en langage naturel placées entre crochets au niveau du mot. Il gère le dialogue à plusieurs locuteurs et le streaming, et ses poids sont publiés pour la recherche et l'usage non commercial.","sortie":"2026-03-09","taille":"4 milliards de paramètres pour l'AR lent, plus 400 millions pour l'AR rapide","langues":{"annoncees":{"annoncees":"plus de 80 langues, dont le français","nombre":80,"plus":true,"francais":true},"prises_en_charge_selon_open_tts":83,"evaluees":["fr","en"]},"preuves":[{"appuie":["sortie"],"lu":"https://fish.audio/blog/fish-audio-open-sources-s2/ (« Mar 9, 2026 · Research Fish Audio Open-Sources S2: Fine-Grained Control Meets Production Streaming »), 17/10/2026"},{"appuie":["langues","presentation"],"lu":"https://huggingface.co/fishaudio/s2-pro/raw/main/README.md (« Trained on over 10M+ hours of audio data across 80+ languages, the system combines reinforcement learning alignment with a dual-autoregressive architecture. »), 17/10/2026"},{"appuie":["langues"],"lu":"https://huggingface.co/fishaudio/s2-pro/raw/main/README.md (« Korean (ko), Spanish (es), Portuguese (pt), Arabic (ar), Russian (ru), French (fr), German (de) »), 17/10/2026"},{"appuie":["taille"],"lu":"https://huggingface.co/fishaudio/s2-pro/raw/main/README.md (« **Slow AR** (4B parameters): Operates along the time axis and predicts the primary semantic codebook. »), 17/10/2026"},{"appuie":["taille"],"lu":"https://huggingface.co/fishaudio/s2-pro/raw/main/README.md (« **Fast AR** (400M parameters): Generates the remaining 9 residual codebooks at each time step »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://huggingface.co/fishaudio/s2-pro/raw/main/README.md (« This model is licensed under the [Fish Audio Research License](LICENSE.md). Research and non-commercial use is permitted free of charge. »), 17/10/2026"}],"classements":[{"id":"synthese/francais","nom":"Synthèse vocale en français","rang":26,"sur":31,"valeur":1049.49,"unite":"Elo","bas":1028.49,"haut":1070.49,"votes":1158,"chiffres":{"prix_par_million_caracteres":15,"date_sortie":"2026-03-10"},"voisins":["boson-ai-higgs-tts-3-4b","coqui-xtts-v2"],"source":{"nom":"Artificial Analysis","mention":"Source : Artificial Analysis","url":"https://artificialanalysis.ai/text-to-speech/model-families/fishaudio","releve":"2026-10-19"}},{"id":"synthese/prononciation","nom":"Prononciation du français","rang":2,"sur":16,"valeur":5.35,"unite":"%","chiffres":{"rtfx":0.4,"taille_milliards":4.95,"langues":83,"clonage":true},"voisins":["k2-fsa-omnivoice","kokoro-82m-v1-0"],"source":{"nom":"Open TTS Leaderboard","mention":"Source : Open TTS Leaderboard, Hugging Face","url":"https://huggingface.co/spaces/hf-audio/open_tts_leaderboard","releve":"2026-10-19"}},{"id":"synthese/artificial-analysis","nom":"Synthèse vocale en anglais, Artificial Analysis","rang":35,"sur":96,"valeur":1113.78,"unite":"Elo","bas":1101.78,"haut":1125.78,"votes":2621,"chiffres":{"prix_par_million_caracteres":15,"date_sortie":"2026-03-10"},"voisins":["speechifyai-simba-3-0","minimax-speech-02-hd"],"source":{"nom":"Artificial Analysis","mention":"Source : Artificial Analysis","url":"https://artificialanalysis.ai/text-to-speech/model-families/fishaudio","releve":"2026-10-19"}}],"noms_dans_les_sources":{"artificial_analysis":"Fish Audio S2 Pro","hf_audio":"fishaudio/s2-pro"}}