今天下午，知名人工智能企业科大讯飞发布了一款名为“星火语音大模型”的多语种语音合成技术。这一技术在首批37个主流语种上的表现已经整体超越了OpenAI公司推出的Whisper-large-v3。
作为我国智能语音技术的领军企业，科大讯飞此次发布的星火语音大模型不仅在技术上取得了重大突破，还在实际应用中展示了其强大的性能。根据发布会上的介绍，星火语音大模型在首批40个语种上的平均MOS分绝对提升了0.25，MOS分达到了4.5，拟人度更是高达83%以上。
这一突破再次证明了我国在人工智能领域的实力，科大讯飞继续保持智能语音技术的国际领先水平。星火语音大模型的发布，不仅为我国人工智能产业的发展注入了新的活力，也为全球多语种语音合成技术的进步做出了重要贡献。
英文标题Title：AI Starlight Voice Model by iFLYTEK Surpasses OpenAI’s Whispr
英文关键词Keywords：iFLYTEK, Starlight Voice, International Leadership
英文新闻内容News content：
This afternoon, iFLYTEK, a well-known artificial intelligence company, released a multilingual speech synthesis technology named “Starlight Voice Model”. This technology has overall surpassed OpenAI’s Whisper-large-v3 in the first batch of 37 mainstream languages.
As a leading company in China’s intelligent voice technology field, iFLYTEK’s Starlight Voice Model has not only achieved a significant breakthrough in technology but also demonstrated its strong performance in practical applications. According to the introduction at the press conference, the Starlight Voice Model has increased the average MOS score by 0.25 in the first batch of 40 languages, reaching a MOS score of 4.5, and a human-like degree of over 83%.
This breakthrough has once again proven the strength of China in the field of artificial intelligence. iFLYTEK continues to maintain its international leading position in intelligent voice technology. The release of the Starlight Voice Model not only injects new vitality into the development of China’s artificial intelligence industry but also makes an important contribution to the progress of global multilingual speech synthesis technology.
【来源】https://news.mydrivers.com/1/961/961266.htm
Views: 1