Huawei machine learning service speech recognition function enables applications to paint "sound" and color

With people's pursuit of the ritual sense of life , Mobile devices 、 Wearable device 、 Smart home devices 、 Car infotainment systems are becoming more and more popular . In these applications , mouse 、 The keyboard is no longer convenient , And voice is the most natural way of communication between human beings , Speech recognition technology seems to have become a major application “ Standard configuration ”. Speech recognition scenarios are widely used , Such as voice input method 、 Voice search 、 Real time subtitles 、 Entertainment 、 Social chat 、 human-computer interaction 、 Driving mode, etc . therefore ,App The integrated speech recognition function can not only free your hands , You can also get a good human-computer interaction experience .

1. Business profile

HMS Core Machine learning services Real time speech recognition service Support short voice input in real time （ Duration not exceeding 60 second ） Convert to text , Real time recognition 60 Speech in seconds . The service uses industry-leading deep learning technology , With the continuous iteration of algorithm and data , At present, the recognition accuracy in the general ideal environment can reach 95% above . At present, Chinese Putonghua is supported （ Including mixed Chinese and English ）、 English 、 French 、 German 、 Spanish 、 Italian 、 The Arabic language 、 Russian 、 Thai, 、 Malay Language 、 Recognition of Filipino .

DEMO demonstration

2 . Scenario introduction

HMS Core Machine learning services Real time speech recognition service It covers many fields in daily life and work , And deep optimization of shopping search 、 Movie search 、 Music search, navigation and other scene recognition capabilities , Further improve the recognition accuracy of these scenes . Using the shopping class App When searching for products , The product name or feature described by voice can be recognized as text to search for the target product . Again , In the use of music App when , You can identify the name or singer of a voice input song as text and search for songs . in addition , When it is not convenient for the driver to input text during driving , The input voice can be converted into text and then the destination can be searched , Make driving safer .

3 . Functional characteristics

• Support real-time word output

• Provides a pickup interface 、 There are two ways without pickup interface

• Support endpoint detection , Accurate positioning of start and end points

• Support mute detection , The voiceless part of the voice does not send voice packets

• Support intelligent conversion of digital format , For example, voice input “ In 2020 ” when , It can be intelligently identified as “2020 year ”.

How to access Huawei machine learning services ？

Huawei The official website of machine learning service Provide you with detailed Document guidance .

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 ~