今日下午，知名人工智能企业科大讯飞举行了一次重大发布会，宣布推出全新星火语音大模型。该模型由多语种语音合成技术支持，在首批37个主流语种上的表现已全面超越OpenAI公司推出的Whisper-large-v3，再次证明了科大讯飞在智能语音技术领域的国际领先地位。
值得一提的是，星火语音大模型不仅在首批37个主流语种上取得突破，还在首批40个语种的平均MOS分上实现了绝对提升0.25，达到了4.5的优异水平。同时，该模型的拟人度也高达83%以上，为用户带来了更为自然流畅的语音体验。
此次星火语音大模型的成功发布，标志着我国在人工智能领域取得了重要突破，科大讯飞将继续致力于推动我国人工智能事业的发展，为全球用户提供更优质、更智能的语音服务。
英文标题Title：New Starfire Voice Model by iFLYTEK Sets New Records
英文关键词Keywords：iFLYTEK, Starfire Voice, International Leadership
英文新闻内容News content：
This afternoon, iFLYTEK, a well-known artificial intelligence company, held a significant press conference where they announced the launch of the new Starfire Voice large model. Supported by multilingual speech synthesis technology, the model has achieved comprehensive superiority over OpenAI’s Whisper-large-v3 in the first batch of 37 mainstream languages. This再次证明s iFLYTEK’s leading position in the field of intelligent speech technology on an international scale.
It is worth mentioning that in addition to the breakthrough in the first batch of 37 mainstream languages, the Starfire Voice large model has also achieved an absolute improvement of 0.25 in the average MOS score of the first batch of 40 languages, reaching an excellent level of 4.5. Moreover, the model’s humanlikeness is as high as 83%, providing users with a more natural and smooth speech experience.
The successful release of the Starfire Voice large model represents an important breakthrough in the field of artificial intelligence in our country. iFLYTEK will continue to work tirelessly to promote the development of artificial intelligence in China and provide users around the world with higher-quality, smarter speech services.
【来源】https://news.mydrivers.com/1/961/961266.htm
Views: 1