{"@context":"https://schema.org","@type":"NewsArticle","generatedAt":"2026-07-23T06:40:50.084Z","headline":"阿里千问发布语音合成大模型 Qwen-Audio-3.0-TTS","description":"阿里千问发布语音合成大模型 Qwen-Audio-3.0-TTS，包含首包延时300ms级别的Flash版本和面向高质量生成的Plus版本，后者在Artificial Analysis榜单夺冠。新模型支持文本嵌入【gasp】等标签调控语气，音频输出提升至48KHz，覆盖16种语言及20种方言，现已全面开放。","url":"https://www.aioga.com/news/cmrt1lhtq0fifbitl3fdq8ykc/","mainEntityOfPage":"https://www.aioga.com/news/cmrt1lhtq0fifbitl3fdq8ykc/","datePublished":"2026-07-20T08:55:05.000Z","dateModified":"2026-07-20T08:55:05.000Z","inLanguage":"zh-CN","publisher":{"@type":"NewsMediaOrganization","name":"Aioga","url":"https://www.aioga.com"},"citation":["https://www.ithome.com/0/979/128.htm","https://aihot.virxact.com/items/cmrt1lhtq0fifbitl3fdq8ykc"],"canonicalUrl":"https://www.aioga.com/news/cmrt1lhtq0fifbitl3fdq8ykc/","directAnswer":{"@type":"Answer","text":"Aioga 编辑摘要：阿里千问发布语音合成大模型 Qwen-Audio-3.0-TTS，包含首包延时300ms级别的Flash版本和面向高质量生成的Plus版本，后者在Artificial Analysis榜单夺冠。 Aioga 将其归入「模型更新」方向，重点关注它对真实使用和行业竞争的影响。","url":"https://www.aioga.com/news/cmrt1lhtq0fifbitl3fdq8ykc/","dateCreated":"2026-07-20T08:55:05.000Z","author":{"@type":"Organization","@id":"https://www.aioga.com/authors/aioga-editorial/#editorial-team","name":"Aioga Editorial Team","url":"https://www.aioga.com/authors/aioga-editorial/"}},"evidence":[{"@type":"CreativeWork","name":"IT之家 source article","url":"https://www.ithome.com/0/979/128.htm","datePublished":"2026-07-20T08:55:05.000Z","provider":{"@type":"Organization","name":"IT之家","url":"https://www.ithome.com/0/979/128.htm"}},{"@type":"CreativeWork","name":"AIHot archive record","url":"https://aihot.virxact.com/items/cmrt1lhtq0fifbitl3fdq8ykc","datePublished":"2026-07-20T08:55:05.000Z","provider":{"@type":"Organization","name":"AIHot","url":"https://aihot.virxact.com/items/cmrt1lhtq0fifbitl3fdq8ykc"}}],"aggregationSource":"IT之家（RSS）","originalPublisher":{"name":"IT之家","url":"https://www.ithome.com/0/979/128.htm"},"article":{"id":"cmrt1lhtq0fifbitl3fdq8ykc","slug":"cmrt1lhtq0fifbitl3fdq8ykc","url":"https://www.aioga.com/news/cmrt1lhtq0fifbitl3fdq8ykc/","title":"阿里千问发布语音合成大模型 Qwen-Audio-3.0-TTS","title_en":"位列 Artificial Analysis 榜首，阿里千问发布语音合成大模型 Qwen-Audio-3.0-TTS","summary":"阿里千问发布语音合成大模型 Qwen-Audio-3.0-TTS，包含首包延时300ms级别的Flash版本和面向高质量生成的Plus版本，后者在Artificial Analysis榜单夺冠。新模型支持文本嵌入【gasp】等标签调控语气，音频输出提升至48KHz，覆盖16种语言及20种方言，现已全面开放。","source":"IT之家（RSS）","sourceUrl":"https://www.ithome.com/0/979/128.htm","aiHotUrl":"https://aihot.virxact.com/items/cmrt1lhtq0fifbitl3fdq8ykc","publishedAt":"2026-07-20T08:55:05.000Z","category":"模型更新","score":68,"selected":false,"articleBody":["IT之家：https://www.ithome.com/ 7 月 20 日消息，阿里千问今日发布语音合成大模型 Qwen-Audio-3.0-TTS，包含面向实时交互的 Flash 版本（首包延时 300ms 级别）和面向高质量生成的 Plus 版本。","官方表示，新模型在细粒度标签控制、freestyle 指令遵循、多语种与方言覆盖以及复杂声学鲁棒性等方面实现系统性提升，让合成语音从“能说话”走向“会表达”。在全球第三方权威榜单 Artificial Analysis，Qwen-Audio-3.0-TTS-Plus 斩获冠军 。","据介绍，Qwen-Audio-3.0-TTS 支持在文本中嵌入 结构化标签 ，用户可直接在文本里嵌入 [gasp]（抽气）、[giggles]（轻笑）、[angry]（愤怒）这类标记，直接对语气、情绪和 breath 类细节进行精准调控。","此外，该模型配套开箱即用的 精品音色库 ，将模型在多语种、多方言、指令控制和细粒度表达上的能力沉淀为可直接调用的音色资源，将陆续上架指令风格音色、20 种方言音色、细粒度控制音色以及 14+ 小语种音色。并将音频输出从提升 24KHz 到 48KHz，相当于从电话和普通语音助手的品质提升到录音棚、影视配音的规格， 单次语音合成可达 3 分钟 。","面向全球多语种场景，Qwen-Audio-3.0-TTS-Plus 进行了专项优化，覆盖中、英、日、韩、德等 16 种语言 ，新增阿拉伯、越南、马来、菲律宾语；并支持广东、重庆、东北、陕西、上海、四川、云南等 20 种方言 。","Plus 版本： https://bailian.console.aliyun.com/cn-beijing ?tab=model# /model-market/detail/qwen-audio-3.0-tts-plus ：https://bailian.console.aliyun.com/cn-beijing?tab=model#/model-market/detail/qwen-audio-3.0-tts-plus","Flash 版本： https://bailian.console.aliyun.com/cn-beijing ?tab=model# /model-market/detail/qwen-audio-3.0-tts-flash ：https://bailian.console.aliyun.com/cn-beijing?tab=model#/model-market/detail/qwen-audio-3.0-tts-flash"],"articleImages":[{"sourceUrl":"https://img.ithome.com/newsuploadfiles/2026/7/1efd8b04-4b44-4385-aefa-c799dee50aa1.png","alt":"","afterParagraph":1,"url":"/media/articles/cmrt1lhtq0fifbitl3fdq8ykc/ecfed73332c064dc.png"},{"sourceUrl":"https://img.ithome.com/newsuploadfiles/2026/7/12f6da56-cda3-408a-827e-924072b8dcfa.png?x-bce-process=image/format,f_auto","alt":"","afterParagraph":4,"url":"/media/articles/cmrt1lhtq0fifbitl3fdq8ykc/7cecf2ef1d0037c6.png"}],"mediaStatus":"ok","articleBodyZh":[],"translationStatus":"","bodyOrigin":"source-page","editorial":{"summary":"Aioga 编辑摘要：阿里千问发布语音合成大模型 Qwen-Audio-3.0-TTS，包含首包延时300ms级别的Flash版本和面向高质量生成的Plus版本，后者在Artificial Analysis榜单夺冠。 Aioga 将其归入「模型更新」方向，重点关注它对真实使用和行业竞争的影响。","background":"背景分析：模型与研究类动态需要结合能力边界、开放方式、成本、可用性和真实任务表现判断，单项指标领先不等于已经形成稳定采用。","viewpoint":"Aioga 判断：这条动态更适合作为行业观察信号，当前信息足以建立线索，但不足以推导长期结论。","implications":"影响分析：对相关团队而言，短期应先核对来源、可用范围和实际成本，再判断是否值得接入或跟进。","nextStep":"后续观察：继续观察官方文档、实际可用性、价格变化、开发者反馈和竞品回应。","evidenceRefs":["title","summary","articleBody"],"confidence":"medium","status":"published","aiGenerated":false,"autoApproved":true,"generatedBy":"rule-safe-fallback","generatedAt":"2026-07-23T06:49:19.110Z","sourceHash":"517b56e863c9b3b4","validation":{"passed":true,"mode":"rule-safe-fallback","checks":["schema","length","source-attribution","no-html"]}},"tags":["模型更新","IT之家（RSS）"],"translations":{"zh-CN":{"title":"阿里千问发布语音合成大模型 Qwen-Audio-3.0-TTS","summary":"阿里千问发布语音合成大模型 Qwen-Audio-3.0-TTS，包含首包延时300ms级别的Flash版本和面向高质量生成的Plus版本，后者在Artificial Analysis榜单夺冠。新模型支持文本嵌入【gasp】等标签调控语气，音频输出提升至48KHz，覆盖16种语言及20种方言，现已全面开放。","category":"模型更新","source":"IT之家","aggregationSource":"IT之家（RSS）","pageTitle":"阿里千问发布语音合成大模型 Qwen-Audio-3.0-TTS - Aioga AI资讯","description":"阿里千问发布语音合成大模型 Qwen-Audio-3.0-TTS，包含首包延时300ms级别的Flash版本和面向高质量生成的Plus版本，后者在Artificial Analysis榜单夺冠。新模型支持文本嵌入【gasp】等标签调控语气，音频输出提升至48KHz，覆盖16种语言及20种方言，现已全面开放。","url":"https://www.aioga.com/news/cmrt1lhtq0fifbitl3fdq8ykc/"},"en":{"title":"Alibaba Qianwen releases speech synthesis large model Qwen-Audio-3.0-TTS","summary":"Alibaba Qianwen has released the Qwen-Audio-3.0-TTS large speech synthesis model, which includes a Flash version with a first-packet delay of about 300ms and a Plus version aimed at high-quality generation, the latter having taken first place on the Artificial Analysis leaderboard. The new model supports controlling tone with tags such as text embedding 【gasp】, improves audio output to 48KHz, covers 16 languages and 20 dialects, and is now fully open.","category":"Models","source":"IT之家（RSS）","aggregationSource":"IT之家（RSS）","pageTitle":"Alibaba Qianwen releases speech synthesis large model Qwen-Audio-3.0-TTS - Aioga AI News","description":"Alibaba Qianwen has released the Qwen-Audio-3.0-TTS large speech synthesis model, which includes a Flash version with a first-packet delay of about 300ms and a Plus version aimed a...","url":"https://www.aioga.com/en/news/cmrt1lhtq0fifbitl3fdq8ykc/","contentTranslated":true,"sourceHash":"f2bfdbbfe07191b5","translatedAt":"2026-07-22T18:45:05.178Z"},"ja":{"title":"アリ千問が音声合成大型モデル Qwen-Audio-3.0-TTS を発表","summary":"アリババの千問は音声合成大規模モデル Qwen-Audio-3.0-TTS を発表しました。初回パッケージは遅延300msレベルのFlash版と、高品質生成向けのPlus版を含み、後者はArtificial Analysisランキングで1位を獲得しました。新モデルはテキスト埋め込み【gasp】などのタグで語気を調整でき、音声出力は48KHzに向上し、16言語および20の方言をカバーしており、現在全面的に公開されています。","category":"モデル更新","source":"IT之家（RSS）","aggregationSource":"IT之家（RSS）","pageTitle":"アリ千問が音声合成大型モデル Qwen-Audio-3.0-TTS を発表 - Aioga AIニュース","description":"アリババの千問は音声合成大規模モデル Qwen-Audio-3.0-TTS を発表しました。初回パッケージは遅延300msレベルのFlash版と、高品質生成向けのPlus版を含み、後者はArtificial Analysisランキングで1位を獲得しました。新モデルはテキスト埋め込み【gasp】などのタグで語気を調整でき、音声出力は48KHzに向上し、16言語...","url":"https://www.aioga.com/ja/news/cmrt1lhtq0fifbitl3fdq8ykc/","contentTranslated":true,"sourceHash":"f2bfdbbfe07191b5","translatedAt":"2026-07-22T18:45:07.455Z"},"ko":{"title":"알리천문, 음성 합성 대형 모델 Qwen-Audio-3.0-TTS 발표","summary":"알리천문은 음성 합성 대형 모델 Qwen-Audio-3.0-TTS를 출시했으며, 첫 패키지 지연 300ms 수준의 Flash 버전과 고품질 생성을 위한 Plus 버전을 포함하고 있다. 후자는 Artificial Analysis 순위에서 1위를 차지했다. 새로운 모델은 텍스트 임베딩 【gasp】 등 태그로 어조 조절을 지원하며, 오디오 출력은 48KHz로 향상되고, 16개 언어와 20개 방언을 커버하며 현재 전면 개방되어 있다.","category":"모델 업데이트","source":"IT之家（RSS）","aggregationSource":"IT之家（RSS）","pageTitle":"알리천문, 음성 합성 대형 모델 Qwen-Audio-3.0-TTS 발표 - Aioga AI 뉴스","description":"알리천문은 음성 합성 대형 모델 Qwen-Audio-3.0-TTS를 출시했으며, 첫 패키지 지연 300ms 수준의 Flash 버전과 고품질 생성을 위한 Plus 버전을 포함하고 있다. 후자는 Artificial Analysis 순위에서 1위를 차지했다. 새로운 모델은 텍스트 임베딩 【gasp】 등 태그로 어조 조절을 지원...","url":"https://www.aioga.com/ko/news/cmrt1lhtq0fifbitl3fdq8ykc/","contentTranslated":true,"sourceHash":"f2bfdbbfe07191b5","translatedAt":"2026-07-22T18:45:57.562Z"},"es":{"title":"Alibaba Qianwen lanza modelo de síntesis de voz Qwen-Audio-3.0-TTS","summary":"Ali Qianwen lanzó el gran modelo de síntesis de voz Qwen-Audio-3.0-TTS, que incluye la versión Flash con un retraso inicial de nivel 300 ms y la versión Plus orientada a la generación de alta calidad, esta última ganadora en la lista Artificial Analysis. El nuevo modelo admite el control del tono mediante etiquetas como 【gasp】 en la incrustación de texto, la salida de audio se ha mejorado a 48 KHz, cubre 16 idiomas y 20 dialectos, y ahora está completamente disponible.","category":"Modelos","source":"IT之家（RSS）","aggregationSource":"IT之家（RSS）","pageTitle":"Alibaba Qianwen lanza modelo de síntesis de voz Qwen-Audio-3.0-TTS - Aioga Noticias de IA","description":"Ali Qianwen lanzó el gran modelo de síntesis de voz Qwen-Audio-3.0-TTS, que incluye la versión Flash con un retraso inicial de nivel 300 ms y la versión Plus orientada a la generac...","url":"https://www.aioga.com/es/news/cmrt1lhtq0fifbitl3fdq8ykc/","contentTranslated":true,"sourceHash":"f2bfdbbfe07191b5","translatedAt":"2026-07-22T18:45:51.328Z"},"fr":{"title":"Alibaba Qianwen publie le grand modèle de synthèse vocale Qwen-Audio-3.0-TTS","summary":"Alibaba Qianwen a lancé le grand modèle de synthèse vocale Qwen-Audio-3.0-TTS, comprenant une version Flash avec un premier paquet retardé de l'ordre de 300 ms et une version Plus destinée à une génération de haute qualité, cette dernière étant numéro un du classement Artificial Analysis. Le nouveau modèle prend en charge les balises de régulation de ton comme 【gasp】 lors de l’intégration de texte, la sortie audio est améliorée à 48 kHz, couvrant 16 langues et 20 dialectes, et est désormais entièrement ouverte.","category":"Modèles","source":"IT之家（RSS）","aggregationSource":"IT之家（RSS）","pageTitle":"Alibaba Qianwen publie le grand modèle de synthèse vocale Qwen-Audio-3.0-TTS - Aioga Actualités IA","description":"Alibaba Qianwen a lancé le grand modèle de synthèse vocale Qwen-Audio-3.0-TTS, comprenant une version Flash avec un premier paquet retardé de l'ordre de 300 ms et une version Plus...","url":"https://www.aioga.com/fr/news/cmrt1lhtq0fifbitl3fdq8ykc/","contentTranslated":true,"sourceHash":"f2bfdbbfe07191b5","translatedAt":"2026-07-22T18:46:46.115Z"},"de":{"title":"Alibaba Qianwen veröffentlicht großes Sprachsynthese-Modell Qwen-Audio-3.0-TTS","summary":"Alibaba Qianwen hat das Sprachsynthese-Großmodell Qwen-Audio-3.0-TTS veröffentlicht, das eine Flash-Version mit erstem Paketverzögerung im 300-ms-Bereich und eine auf hochwertige Erzeugung ausgerichtete Plus-Version enthält, wobei letztere die Spitzenposition in der Artificial Analysis-Rangliste erreicht hat. Das neue Modell unterstützt die Steuerung des Sprachtons durch Tags wie 【gasp】 für Texteingaben, die Audioausgabe wurde auf 48 kHz verbessert, es deckt 16 Sprachen und 20 Dialekte ab und ist nun vollständig verfügbar.","category":"模型更新","source":"IT之家（RSS）","aggregationSource":"IT之家（RSS）","pageTitle":"Alibaba Qianwen veröffentlicht großes Sprachsynthese-Modell Qwen-Audio-3.0-TTS - Aioga KI-News","description":"Alibaba Qianwen hat das Sprachsynthese-Großmodell Qwen-Audio-3.0-TTS veröffentlicht, das eine Flash-Version mit erstem Paketverzögerung im 300-ms-Bereich und eine auf hochwertige E...","url":"https://www.aioga.com/de/news/cmrt1lhtq0fifbitl3fdq8ykc/","contentTranslated":true,"sourceHash":"f2bfdbbfe07191b5","translatedAt":"2026-07-22T18:46:39.128Z"},"pt-BR":{"title":"Ali Qianwen lança grande modelo de síntese de voz Qwen-Audio-3.0-TTS","summary":"A Ali Qianwen lançou o grande modelo de síntese de voz Qwen-Audio-3.0-TTS, que inclui a versão Flash com latência de 300ms no primeiro pacote e a versão Plus voltada para geração de alta qualidade, esta última conquistou o primeiro lugar no ranking Artificial Analysis. O novo modelo suporta a regulação de entonação com tags como 【gasp】, a saída de áudio foi aprimorada para 48KHz, cobre 16 idiomas e 20 dialetos, e agora está totalmente disponível.","category":"模型更新","source":"IT之家（RSS）","aggregationSource":"IT之家（RSS）","pageTitle":"Ali Qianwen lança grande modelo de síntese de voz Qwen-Audio-3.0-TTS - Aioga Notícias de IA","description":"A Ali Qianwen lançou o grande modelo de síntese de voz Qwen-Audio-3.0-TTS, que inclui a versão Flash com latência de 300ms no primeiro pacote e a versão Plus voltada para geração d...","url":"https://www.aioga.com/pt-BR/news/cmrt1lhtq0fifbitl3fdq8ykc/","contentTranslated":true,"sourceHash":"f2bfdbbfe07191b5","translatedAt":"2026-07-22T18:47:26.965Z"},"ru":{"title":"Али Цяньвэнь выпустил большую модель синтеза речи Qwen-Audio-3.0-TTS","summary":"Али Цяньвэнь выпустила крупную модель синтеза речи Qwen-Audio-3.0-TTS, включающую Flash-версию с задержкой первого пакета на уровне 300 мс и Plus-версию для высококачественной генерации, которая заняла первое место в рейтинге Artificial Analysis. Новая модель поддерживает регулировку интонации с помощью тегов, таких как текстовое встраивание 【gasp】, выход аудио увеличен до 48 КГц, охватывает 16 языков и 20 диалектов, и теперь полностью доступна.","category":"模型更新","source":"IT之家（RSS）","aggregationSource":"IT之家（RSS）","pageTitle":"Али Цяньвэнь выпустил большую модель синтеза речи Qwen-Audio-3.0-TTS - Aioga Новости ИИ","description":"Али Цяньвэнь выпустила крупную модель синтеза речи Qwen-Audio-3.0-TTS, включающую Flash-версию с задержкой первого пакета на уровне 300 мс и Plus-версию для высококачественной гене...","url":"https://www.aioga.com/ru/news/cmrt1lhtq0fifbitl3fdq8ykc/","contentTranslated":true,"sourceHash":"f2bfdbbfe07191b5","translatedAt":"2026-07-22T18:47:30.118Z"},"ar":{"title":"ألي تشيان وِن تُصدر نموذجًا كبيرًا لتوليد الصوت Qwen-Audio-3.0-TTS","summary":"أعلنت شركة علي تشيان ون عن نموذج كبير لتوليد الصوت Qwen-Audio-3.0-TTS، ويشمل النسخة الفلاشية ذات التأخير الأولي بمقدار 300 مللي ثانية ونسخة بلس المخصصة لإنتاج عالي الجودة، والتي فازت بالمركز الأول في قائمة Artificial Analysis. النموذج الجديد يدعم إدراج النصوص مع علامات مثل 【gasp】 للتحكم في النبرة، وتم تحسين جودة الصوت لتصل إلى 48 كيلو هرتز، ويغطي 16 لغة و20 لهجة، وهو متاح الآن بشكل كامل.","category":"模型更新","source":"IT之家（RSS）","aggregationSource":"IT之家（RSS）","pageTitle":"ألي تشيان وِن تُصدر نموذجًا كبيرًا لتوليد الصوت Qwen-Audio-3.0-TTS - Aioga أخبار الذكاء الاصطناعي","description":"أعلنت شركة علي تشيان ون عن نموذج كبير لتوليد الصوت Qwen-Audio-3.0-TTS، ويشمل النسخة الفلاشية ذات التأخير الأولي بمقدار 300 مللي ثانية ونسخة بلس المخصصة لإنتاج عالي الجودة، والتي فا...","url":"https://www.aioga.com/ar/news/cmrt1lhtq0fifbitl3fdq8ykc/","contentTranslated":true,"sourceHash":"f2bfdbbfe07191b5","translatedAt":"2026-07-22T18:48:17.154Z"},"hi":{"title":"अली क़ियानवेन ने वॉइस सिंथेसिस बड़ा मॉडल Qwen-Audio-3.0-TTS जारी किया","summary":"अली क़ियानवेन ने वॉइस सिंथेसिस बड़े मॉडल Qwen-Audio-3.0-TTS जारी किया, जिसमें शुरुआती पैकेज की विलंबता 300ms स्तर का फ्लैश संस्करण और उच्च गुणवत्ता वाली उत्पादन के लिए प्लस संस्करण शामिल है, जिसमें बाद वाले ने Artificial Analysis सूची में पहला स्थान प्राप्त किया। नया मॉडल टेक्स्ट एम्बेडिंग [gasp] जैसी टैग्स के माध्यम से स्वर नियंत्रित करने का समर्थन करता है, ऑडियो आउटपुट को 48KHz तक बढ़ाया गया है, 16 भाषाओं और 20 बोलियों को कवर करता है, और अब यह पूरी तरह से खुला है।","category":"模型更新","source":"IT之家（RSS）","aggregationSource":"IT之家（RSS）","pageTitle":"अली क़ियानवेन ने वॉइस सिंथेसिस बड़ा मॉडल Qwen-Audio-3.0-TTS जारी किया - Aioga AI समाचार","description":"अली क़ियानवेन ने वॉइस सिंथेसिस बड़े मॉडल Qwen-Audio-3.0-TTS जारी किया, जिसमें शुरुआती पैकेज की विलंबता 300ms स्तर का फ्लैश संस्करण और उच्च गुणवत्ता वाली उत्पादन के लिए प्लस संस्करण...","url":"https://www.aioga.com/hi/news/cmrt1lhtq0fifbitl3fdq8ykc/","contentTranslated":true,"sourceHash":"f2bfdbbfe07191b5","translatedAt":"2026-07-22T18:48:32.213Z"},"it":{"title":"Alibaba Qianwen rilascia il grande modello di sintesi vocale Qwen-Audio-3.0-TTS","summary":"Ali Qianwen ha rilasciato il grande modello di sintesi vocale Qwen-Audio-3.0-TTS, che include una versione Flash a bassa latenza di 300 ms e una versione Plus orientata alla generazione di alta qualità, quest'ultima vincitrice della classifica Artificial Analysis. Il nuovo modello supporta il controllo dell'intonazione tramite tag come 【gasp】 per gli embedding testuali, l'uscita audio è stata migliorata a 48 kHz, copre 16 lingue e 20 dialetti, ed è ora completamente disponibile.","category":"模型更新","source":"IT之家（RSS）","aggregationSource":"IT之家（RSS）","pageTitle":"Alibaba Qianwen rilascia il grande modello di sintesi vocale Qwen-Audio-3.0-TTS - Aioga Notizie IA","description":"Ali Qianwen ha rilasciato il grande modello di sintesi vocale Qwen-Audio-3.0-TTS, che include una versione Flash a bassa latenza di 300 ms e una versione Plus orientata alla genera...","url":"https://www.aioga.com/it/news/cmrt1lhtq0fifbitl3fdq8ykc/","contentTranslated":true,"sourceHash":"f2bfdbbfe07191b5","translatedAt":"2026-07-22T18:49:18.091Z"},"nl":{"title":"Alibaba Qianwen brengt spraakgeneratie groot model Qwen-Audio-3.0-TTS uit","summary":"Alibaba Qianwen heeft het spraaksynthesemodel Qwen-Audio-3.0-TTS uitgebracht, dat een Flash-versie met een initiële vertraging van 300 ms en een Plus-versie voor hoogwaardige generatie bevat, waarbij de laatstgenoemde de eerste plaats behaalde in de Artificial Analysis-ranglijst. Het nieuwe model ondersteunt toonregeling via labels zoals tekstinbedding 【gasp】, de audio-uitvoer is verhoogd naar 48 kHz, en het dekt 16 talen en 20 dialecten. Het is nu volledig beschikbaar.","category":"模型更新","source":"IT之家（RSS）","aggregationSource":"IT之家（RSS）","pageTitle":"Alibaba Qianwen brengt spraakgeneratie groot model Qwen-Audio-3.0-TTS uit - Aioga AI-nieuws","description":"Alibaba Qianwen heeft het spraaksynthesemodel Qwen-Audio-3.0-TTS uitgebracht, dat een Flash-versie met een initiële vertraging van 300 ms en een Plus-versie voor hoogwaardige gener...","url":"https://www.aioga.com/nl/news/cmrt1lhtq0fifbitl3fdq8ykc/","contentTranslated":true,"sourceHash":"f2bfdbbfe07191b5","translatedAt":"2026-07-22T18:49:22.467Z"},"tr":{"title":"Alibaba Qianwen, ses sentezi büyük modeli Qwen-Audio-3.0-TTS'yi yayınladı","summary":"Alibaba Qianwen, ses sentezi büyük modeli Qwen-Audio-3.0-TTS'yi yayınladı; model, ilk paket gecikmesi 300ms seviyesinde olan Flash sürümü ve yüksek kaliteli üretim için Plus sürümünü içeriyor. Plus sürüm, Artificial Analysis sıralamasında birinci oldu. Yeni model, metin gömme 【gasp】 gibi etiketlerle tonlama kontrolünü destekliyor, ses çıktısı 48KHz'e yükseltilmiş, 16 dil ve 20 lehçeyi kapsıyor ve artık tamamen erişime açıldı.","category":"模型更新","source":"IT之家（RSS）","aggregationSource":"IT之家（RSS）","pageTitle":"Alibaba Qianwen, ses sentezi büyük modeli Qwen-Audio-3.0-TTS'yi yayınladı - Aioga AI Haberleri","description":"Alibaba Qianwen, ses sentezi büyük modeli Qwen-Audio-3.0-TTS'yi yayınladı; model, ilk paket gecikmesi 300ms seviyesinde olan Flash sürümü ve yüksek kaliteli üretim için Plus sürümü...","url":"https://www.aioga.com/tr/news/cmrt1lhtq0fifbitl3fdq8ykc/","contentTranslated":true,"sourceHash":"f2bfdbbfe07191b5","translatedAt":"2026-07-22T18:50:14.665Z"},"vi":{"title":"Alibaba Qianwen phát hành mô hình tổng hợp giọng nói Qwen-Audio-3.0-TTS","summary":"Alibaba Qianwen đã ra mắt mô hình tổng hợp giọng nói lớn Qwen-Audio-3.0-TTS, bao gồm phiên bản Flash có độ trễ gói đầu 300ms và phiên bản Plus hướng đến chất lượng tạo ra cao, phiên bản sau đã giành vị trí quán quân trên bảng xếp hạng Artificial Analysis. Mô hình mới hỗ trợ điều chỉnh ngữ điệu bằng các thẻ như 【gasp】 nhúng trong văn bản, đầu ra âm thanh nâng lên 48KHz, bao phủ 16 ngôn ngữ và 20 phương ngữ, hiện đã mở hoàn toàn.","category":"模型更新","source":"IT之家（RSS）","aggregationSource":"IT之家（RSS）","pageTitle":"Alibaba Qianwen phát hành mô hình tổng hợp giọng nói Qwen-Audio-3.0-TTS - Tin tức AI Aioga","description":"Alibaba Qianwen đã ra mắt mô hình tổng hợp giọng nói lớn Qwen-Audio-3.0-TTS, bao gồm phiên bản Flash có độ trễ gói đầu 300ms và phiên bản Plus hướng đến chất lượng tạo ra cao, phiê...","url":"https://www.aioga.com/vi/news/cmrt1lhtq0fifbitl3fdq8ykc/","contentTranslated":true,"sourceHash":"f2bfdbbfe07191b5","translatedAt":"2026-07-22T18:50:10.849Z"},"id":{"title":"Ali Qianwen merilis model besar sintesis suara Qwen-Audio-3.0-TTS","summary":"Ali Qianwen merilis model besar sintesis suara Qwen-Audio-3.0-TTS, mencakup versi Flash dengan latensi paket pertama sekitar 300ms dan versi Plus untuk generasi berkualitas tinggi, yang terakhir menempati posisi pertama di daftar Artificial Analysis. Model baru mendukung pengaturan intonasi melalui tag seperti 【gasp】 dalam penyematan teks, keluaran audio ditingkatkan hingga 48KHz, mencakup 16 bahasa dan 20 dialek, dan sekarang telah sepenuhnya dibuka.","category":"模型更新","source":"IT之家（RSS）","aggregationSource":"IT之家（RSS）","pageTitle":"Ali Qianwen merilis model besar sintesis suara Qwen-Audio-3.0-TTS - Berita AI Aioga","description":"Ali Qianwen merilis model besar sintesis suara Qwen-Audio-3.0-TTS, mencakup versi Flash dengan latensi paket pertama sekitar 300ms dan versi Plus untuk generasi berkualitas tinggi,...","url":"https://www.aioga.com/id/news/cmrt1lhtq0fifbitl3fdq8ykc/","contentTranslated":true,"sourceHash":"f2bfdbbfe07191b5","translatedAt":"2026-07-22T18:51:01.206Z"},"th":{"title":"อาลีฉีเหวินเปิดตัวโมเดลสังเคราะห์เสียง Qwen-Audio-3.0-TTS","summary":"阿里千问发布เสียงสังเคราะห์โมเดลขนาดใหญ่ Qwen-Audio-3.0-TTS ซึ่งรวมถึงเวอร์ชัน Flash แบบแพ็กเกจแรกที่มีความหน่วงระดับ 300ms และเวอร์ชัน Plus ที่มุ่งเน้นการสร้างคุณภาพสูง รุ่นหลังคว้าแชมป์ในรายการ Artificial Analysis โมเดลใหม่รองรับการฝังข้อความและปรับโทนเสียงด้วยแท็ก เช่น 【gasp】 การส่งออกเสียงปรับปรุงเป็น 48KHz ครอบคลุม 16 ภาษาและ 20 สำเนียง ปัจจุบันเปิดให้ใช้งานเต็มรูปแบบแล้ว","category":"模型更新","source":"IT之家（RSS）","aggregationSource":"IT之家（RSS）","pageTitle":"อาลีฉีเหวินเปิดตัวโมเดลสังเคราะห์เสียง Qwen-Audio-3.0-TTS - ข่าว AI Aioga","description":"阿里千问发布เสียงสังเคราะห์โมเดลขนาดใหญ่ Qwen-Audio-3.0-TTS ซึ่งรวมถึงเวอร์ชัน Flash แบบแพ็กเกจแรกที่มีความหน่วงระดับ 300ms และเวอร์ชัน Plus ที่มุ่งเน้นการสร้างคุณภาพสูง รุ่นหลังคว้าแชมป...","url":"https://www.aioga.com/th/news/cmrt1lhtq0fifbitl3fdq8ykc/","contentTranslated":true,"sourceHash":"f2bfdbbfe07191b5","translatedAt":"2026-07-22T18:51:09.982Z"},"pl":{"title":"Alibaba Qianwen wprowadza dużą model syntezy mowy Qwen-Audio-3.0-TTS","summary":"Alibaba Qianwen wprowadziło duży model syntezy mowy Qwen-Audio-3.0-TTS, który zawiera wersję Flash z opóźnieniem pierwszego pakietu na poziomie 300 ms oraz wersję Plus przeznaczoną do wysokiej jakości generowania, która zdobyła pierwsze miejsce na liście Artificial Analysis. Nowy model obsługuje regulację intonacji tekstem za pomocą tagów takich jak 【gasp】, a jakość dźwięku została zwiększona do 48 kHz, obejmuje 16 języków i 20 dialektów i jest teraz w pełni udostępniony.","category":"模型更新","source":"IT之家（RSS）","aggregationSource":"IT之家（RSS）","pageTitle":"Alibaba Qianwen wprowadza dużą model syntezy mowy Qwen-Audio-3.0-TTS - Aioga Wiadomości AI","description":"Alibaba Qianwen wprowadziło duży model syntezy mowy Qwen-Audio-3.0-TTS, który zawiera wersję Flash z opóźnieniem pierwszego pakietu na poziomie 300 ms oraz wersję Plus przeznaczoną...","url":"https://www.aioga.com/pl/news/cmrt1lhtq0fifbitl3fdq8ykc/","contentTranslated":true,"sourceHash":"f2bfdbbfe07191b5","translatedAt":"2026-07-22T18:51:59.299Z"}}}}