{"rewrite":{"id":"r_d7bc00e95265417bc046766b","clusterId":"c_399308fcf59e152fa66f7b81","slug":"alibaba-s-qwen-audio-3-0-tts-tops-artificial-analysis-ranking","model":"deepseek-v4-flash:free","headline":"Alibaba's Qwen-Audio-3.0-TTS Tops Artificial Analysis Ranking","summary":"Alibaba's Tongyi Lab released Qwen-Audio-3.0-TTS, a text-to-speech model with two variants. The Plus model ranks first on Artificial Analysis's TTS leaderboard. It supports 16 languages including Japanese, offers natural-language control over emotion and pacing, and can clone voices from noisy reference audio.","whyItMatters":"The Plus model's top ranking on an independent TTS leaderboard, combined with Japanese support and improved voice cloning from imperfect audio, positions Alibaba's open release as a direct competitor in the text-to-speech space.","webCardHtml":"\u003cp\u003eTongyi Lab, the Alibaba research unit behind the Qwen series, has released Qwen-Audio-3.0-TTS, a text-to-speech model that converts input text into speech and can reproduce a specific person\u0026#39;s voice from a reference audio clip.\u003c/p\u003e\u003cp\u003eThe release includes two models. Flash is optimized for real-time interaction, outputting the first audio data in roughly 300 milliseconds. Plus prioritizes naturalness and timbre fidelity over speed, and it has taken first place on Artificial Analysis\u0026#39;s TTS leaderboard, an independent third-party ranking.\u003c/p\u003e\u003cp\u003eThe model supports 16 languages, including Japanese. Users can control emotion, character settings, scenario, and speaking pace through natural-language instructions, and can insert tags directly into the target text to specify non-verbal details like breaths, laughter, and tone shifts. The voice cloning function also improves, automatically suppressing noise from imperfect reference audio while preserving the original voice quality.\u003c/p\u003e","blueskyPost":"Qwen-Audio-3.0-TTS leads the TTS leaderboard, but its real edge is handling noisy reference audio, which makes voice cloning practical outside studio conditions.","twitterPost":"Qwen-Audio-3.0-TTS tops the TTS ranking, but cloning from noisy audio is the feature that widens its use beyond clean studio setups.","threadsPost":"Qwen-Audio-3.0-TTS taking the top spot on Artificial Analysis is notable, but the practical win is its tolerance for noisy reference audio. That lowers the barrier for voice cloning, moving it from a controlled studio task to something feasible with everyday recordings.","newsletterBlurb":"Alibaba's Tongyi Lab released Qwen-Audio-3.0-TTS, a text-to-speech model with two variants. The Plus model ranks first on Artificial Analysis's TTS leaderboard. It supports 16 languages including Japanese, offers natural-language control over emotion and pacing, and can clone voices from noisy reference audio.","attributionJson":"[{\"source\":\"GameBusiness.jp\",\"url\":\"https://www.gamebusiness.jp/article/2026/07/28/27559.html\",\"title\":\"音声合成ランキングで1位を獲得したAI「Qwen-Audio-3.0-TTS」をアリババが公開。日本語対応、音声クローンも可能（生成AIクローズアップ）\"}]","lintFlagsJson":null,"lintHits":0,"costUsd":0,"inputTokens":4945,"outputTokens":655,"status":"published","repairAttempts":0,"nextRepairAt":null,"factsAttemptedAt":1786245961,"createdAt":"2026-08-09T03:18:26.000Z","publishedAt":"2026-08-09T03:21:44.000Z","updatedAt":"2026-08-09T03:18:26.000Z"},"cluster":{"id":"c_399308fcf59e152fa66f7b81","canonicalTitle":"音声合成ランキングで1位を獲得したAI「Qwen-Audio-3.0-TTS」をアリババが公開。日本語対応、音声クローンも可能（生成AIクローズアップ）","representativeArticleId":"a_3dd8e32aff82d45f7ac98e19","sourceCount":1,"writtenSourceCount":1,"writeAttempts":0,"isSolo":true,"entitiesJson":"{\"anime_titles\":[],\"manga_titles\":[],\"work_titles\":[\"Qwen-Audio-3.0-TTS\"],\"studios\":[\"Alibaba\",\"Tongyi Lab\"],\"people\":[],\"type\":\"news\",\"domain\":\"other\",\"is_roundup\":false}","contentType":"news","status":"published","firstSeenAt":"2026-07-27T21:45:03.000Z","lastSeenAt":"2026-07-27T21:45:03.000Z","updatedAt":"2026-08-09T03:21:44.000Z"},"attribution":[{"source":"GameBusiness.jp","url":"https://www.gamebusiness.jp/article/2026/07/28/27559.html","title":"音声合成ランキングで1位を獲得したAI「Qwen-Audio-3.0-TTS」をアリババが公開。日本語対応、音声クローンも可能（生成AIクローズアップ）"}],"entities":{"anime_titles":[],"manga_titles":[],"work_titles":["Qwen-Audio-3.0-TTS"],"studios":["Alibaba","Tongyi Lab"],"people":[],"type":"news","domain":"other","is_roundup":false},"keyFacts":null}