Meta公司近日推出了一款AI声音生成模型AudioBox，能够同时接收语音及文字输入。据称，AudioBox能生成各种环境音、自然对话语音，并整合了音频生成和编辑能力，以便于用户自由生成自己所需的音频。AudioBox基于Meta今年6月推出的Voicebox AI模型，能够支持语音及文字同时输入，并生成多层次声音。用户可以使用语音及文字描述，让AudioBox生成所需的音频。这款模型能够帮助用户在各种环境中生成自然的声音，并整合了音频生成和编辑能力，使用户能够自由生成自己所需的音频。
新闻翻译：
Meta has recently released an AI audio generation model, AudioBox, which can receive both voice and text inputs. According to reports, AudioBox can generate various environmental sounds and natural speech, and it also整合了 audio generation and editing capabilities, allowing users to generate audio as they need. AudioBox is based on the Voicebox AI model, which was launched by Meta in June this year, and it can support both voice and text inputs to generate multiple layers of sound. Users can use voice and text to describe what they need, and AudioBox will generate the audio accordingly. This model can help users generate natural sounds in various environments, and it also整合了 audio generation and editing capabilities, allowing users to generate audio as they need.
【来源】https://www.ithome.com/0/736/876.htm
Views: 1