Primary sourceIT之家
View sourceEvent ·Multi-source
蚂蚁百灵开源 Ming-3.0 语音系列五款模型，覆盖 TTS、音频理解与 ASR
蚂蚁百灵于 10 月 20 日开源 Ming-3.0 语音系列，共五款模型，分为语音合成、音频理解与对话、语音识别三组。Ming-3.0-tts 支持 Voice Design 文字描述音色和导演模式标注台词情绪与重读。Ming-3.0-audio-flash / tiny 可直接对长音频做转写、总结和结构化输出，保留情绪与说话人信息。Ming-3.0-asr-tiny 已用于阿福医生助手、支付宝搜索等蚂蚁内部业务，模型权重在 Hugging Face 和 ModelScope 提供。
Current read
蚂蚁一口气开源了五款语音模型，TTS 还带导演模式，能指定哪句重读、情绪怎么变，直接下载或调 API 都行。
5 sources51° AI heatLast updated 2026/10/20 01:34:53
Evidence
5 sourcesRelated source 1Artificial Analysis
View sourceRelated source 2AIGCLINK
View sourceRelated source 3vLLM
View sourceRelated source 4@atomic_chat_hq
View sourceConflict check
Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.