[Original] Use large models to automatically take notes, extract PPT, and organize interviews. Alibaba Cloud's new AI product 'Tongyi Listening' is open for public testing

Read Chuang/Shenzhen Business Daily reporter Chen Shu

On June 1, Alibaba Cloud announced the progress of the Tongyi large model. The new AI product "Tongyi Listening" focusing on audio and video content was officially launched, becoming the first large model application product in China open to public testing.

Tongyi Tingwu is connected to the understanding and summarization capabilities of Tongyi Qianwen’s large model, and can become a powerful AI assistant for users in work and study, helping to efficiently complete the transcribing, retrieval, summarizing and summarizing of audio and video content anytime and anywhere. Organizing, such as using large models to automatically take notes, organize interviews, extract PPT, etc. During the public beta period, users can receive free transcription time for more than 100 hours of listening comprehension.

"In a different way, audio and video can be easily read, organized and shared." Alibaba Cloud CTO Zhou Jingren introduced that Tingwu is a work-study AI assistant that targets audio and video content scenarios with high added value of knowledge. , such as meetings, classes, interviews, training, interviews, live broadcasts, watching videos, listening to podcasts, etc., can quickly refine and precipitate knowledge through the latest AI technology such as large models.

[caption id="attachment_3846381" align="alignnone" width="1000"]

Tongyi Listening can automatically generate full-text abstracts, chapter summaries, and speech summaries for audio and video[/caption]

Tingwu showed a live demonstration, which included more than a dozen artificial intelligence functions, which can comprehensively improve the efficiency of knowledge transfer from audio and video to graphics and text. In addition to "good hearing", it can generate meeting records with high accuracy and distinguish different speakers. This AI assistant is also "extremely perceptive". The large model can divide audio and video into chapters in one second and form summary, summary and full text of each speaker. Points of view, organization of focus and to-do items. A large model will be launched in the near future, including functions such as one-click extraction of PPT, AI questioning of multiple audio and video contents, and summarizing specific paragraphs.

[caption id="attachment_3846394" align="alignnone" width="736"]

Tongyi Tingwu Chrome plug-in will be available for download to all users in the near future[/caption]

For some niche scenarios, AudioWu has also set up many "treasure functions": open the Chrome plug-in, foreign language learners and hearing-impaired people can use the bilingual floating subtitle bar to watch videos without subtitles anytime and anywhere. When schedule conflicts, listen to Wu can also become a "meeting stand-in" for professionals. The AI ​​can record the meeting and organize key points when the sound is muted. The transcription results can be downloaded as subtitle files to facilitate video post-production by new media practitioners. Review of Q&A organized by Wu It allows reporters, analysts, lawyers, HR and other groups to organize interviews more efficiently.

Tingwu is connected with Alibaba Cloud Disk, and audio and video content on the cloud disk can be transcribed with one click. Tingwu users who registered during the public beta period will also receive larger Alibaba Cloud Disk storage space, and can be online in the cloud disk. Subtitles can also be automatically produced when playing videos.

Zhou Jingren introduced that Tingwu integrates Alibaba’s most advanced voice and language technology. It has built-in Alibaba's new generation industrial-grade speech recognition model, and the recognition accuracy ranks first in multiple authoritative Chinese data sets; it integrates the self-developed speech semantics multi-modal speaker algorithm, and can distinguish roles in scenes where more than 10 people are speaking; After accessing the Tongyi Qianwen large model, it can summarize tens of thousands of words of audio and video content, leading the country in terms of factual accuracy and completeness of key points, and supports accurate question and answer understanding across multiple audio and video contents.

Reviewer: Tan Lugang

The above is the detailed content of [Original] Use large models to automatically take notes, extract PPT, and organize interviews. Alibaba Cloud's new AI product 'Tongyi Listening' is open for public testing. For more information, please follow other related articles on the PHP Chinese website!