当前位置:网站首页>HMS core machine learning service creates a new "sound" state of simultaneous interpreting translation, and AI makes international exchanges smoother

HMS core machine learning service creates a new "sound" state of simultaneous interpreting translation, and AI makes international exchanges smoother

2022-07-06 21:05:00 51CTO

2022 year 6 month ,HMS Core​ ​ Machine learning services ​​ Provide a new open capability for developers —— simultaneous interpretation , adopt AI Voice technology reduces resource costs , Strengthen communication , It aims to help developers create a variety of simultaneous interpretation applications .

HMS Core Simultaneous interpretation covers machine learning services ​ ​ speech recognition 、 translate 、 speech synthesis ​​ And so on , First, convert the input real-time voice into text , Then translate the text into another language , Finally, the translated text is converted into voice for playback . Simultaneous interpretation ability It can help solve cross language real-time communication in a variety of scenarios , Support the translation between Chinese and English , Provide a variety of voice broadcasts , It can be widely used in various fields 、 The environment is complex 、 Real time meetings 、 Live broadcast and other scenes .

HMS Core Machine learning services create a new simultaneous interpreting translation “ The sound ” state ,AI Make international exchanges smoother _ developer

speech recognition + Machine translation , Both quality and efficiency

about ​ ​ simultaneous interpretation ​​ for , Accurate source language input + Translation output is an important measure . In the face of simultaneous interpretation, the main scene —— The conference speech 、 Subtitle live broadcast 、 Meeting interview 、 Wisdom Education …… Often accompanied by long-term continuous input of audio , End to side recognition through speech energy detection 、 Mute detection 、 Heartbeat detection and other algorithms , Realize the effective sentence breaking of long audio , Thus, the effective speech segment is sent to the speech recognition module , Improve the efficiency of speech recognition , At the same time, it reduces the time delay of simultaneous interpretation , Reduce the impact of noise on the recognition effect .

On the other hand , After voice is converted into text , There are some recognition errors 、 Colloquial expression 、 There are many modal particles 、 The repeated expression of some contents leads to the problem that the text is not fluent and the recognition text is not carefully broken . In view of these situations , Machine learning service is used in text processing module NLP Semantic understanding 、 Homophone ambiguity processing 、 Ambient sound processing 、 Colloquial processing and other error correction techniques , To achieve smooth text 、 Automatic sentence breaking and other functions , Make sure to return high-quality translated text , Thus, speech recognition and translation functions are strengthened , Improve the effect of simultaneous interpretation .

A variety of scenes are covered , Bilingual output of Chinese and English subtitles

Simultaneous interpretation can be applied to face-to-face cross language communication , It can also play a role in remote communication . Whether it's face-to-face multilingual meetings 、 Teleconferencing 、 Or watch foreign language videos , Simultaneous interpretation ability can generate bilingual subtitles in real time , Reduce the cost of understanding , Improve work and learning efficiency .

Customized voice broadcast , Listen and watch without mistake

Simultaneous interpretation ability through advanced deep neural synthesis technology , It can output audio stream data in real time , At the same time, it provides a variety of Chinese 、 English male 、 Female voice choice , So that the pronunciation can achieve a more real and natural effect . Reduce delay through real-time voice broadcast , Combined with real-time subtitle content , Provide participants with audio-visual immersion simultaneous interpretation experience .

HMS Core Machine learning services create a new simultaneous interpreting translation “ The sound ” state ,AI Make international exchanges smoother _ simultaneous interpretation _02

Through speech recognition 、 Machine translation 、 The organic integration of the three technologies of speech synthesis ,HMS Core​ ​ Machine learning services ​​ Provide developers with low latency 、 High accuracy of simultaneous interpretation , Help users conduct international exchanges more smoothly , Create new simultaneous interpretation “ The sound ” state . Developers are welcome to log in ​ ​ Machine learning service homepage ​​, Learn more about product introduction and access preparation .

Learn more >>

visit ​ ​ Official website of Huawei developer Alliance ​​ obtain   Development guidance document
Huawei mobile service open source warehouse address : GitHub Gitee

Pay attention to our , The first time to understand HMS Core Latest technical information ~

原网站

版权声明
本文为[51CTO]所创,转载请带上原文链接,感谢
https://yzsam.com/2022/187/202207061233055674.html