WeBank saw the chance to do good to society with AI technology by harnessing big data.

It utilizes deep neural networks to re-create the voice of the parent(s).

Firstly, speech fragments from either the father or mother are recorded.

Then these fragments are fed into the deep neural networks that then analyze the speech patterns and re-create more representations.

Finally, a speech synthesis system is created using the voice of the originator (father or mother). The AI system is then housed in a custom-built speaker.

Remotely controlled by parents via a mobile application, the system is able to tell any story provided the story text is available.

WeBank and Wechat iHearing team created a neural voice cloning system that takes a few audio samples as input.

They used two approaches build their neural cloning system: speaker adaptation and speaker encoding.

Speaker adaptation is based on fine-tuning a multi-speaker generative model with a few cloning samples. Speaker encoding is based on training a separate model to directly infer a new speaker embedding from cloning audios and to be used with a multi-speaker generative model.

In terms of the naturalness of the speech and its similarity to original speaker, both approaches can achieve good performance, even with very few cloning audios.

While speaker adaptation can achieve better naturalness and similarity, the cloning time or required memory for the speaker encoding approach is significantly less, making it favorable for low-resource deployment.

65% percent of the left-children get a stronger emotional connection with their parents. 43% of them get higher scores at school.

WeBank aims to create and distribute 3000 AI speakers in 2018 more across China and create a conversation about the impact of rural-to-urban migration and how it affects the future generation of China.

Online conversation about this issue was seeded on social media by key opinion leaders and WeBank. On Weibo (China’s equivalent of Twitter), people are beginning to express appreciation and empathy to migrant workers. The project was shared on both Weibo and WeChat, gaining traction as time goes. There’s recognition for the relatively new WeBank as a bank that uses its financial clout to innovatively alleviate societal issues.

Most importantly, Hear My Love shines the spotlight on an issue that China forgets - in the rush for development, the rural areas are emptying into urban cities at a cost that China cannot afford to pay.

WeBank is China's most innovative bank which heavily relies on AI technology to drive their business. WeBank is invested by China's tech giant Tencent.

Through big data and deep learning technology, Hear My Love is able to create a speech synthesis system (Text-To-Speech) only by recording a few speech fragments from the left-behind children's parents, normally migrant workers working in big cities.

The speech synthesis system then can take the place of parents to tell stories to their children (living in rural areas), when they are busy making money to give the family a better future.

Collaborating with local teachers and a team of psychologists, WeBank identified children that are the most vulnerable, emotionally. Once identified, WeBank located their parents in the cities where they work, and invited them to a recording session.

We obtained voice data from them which is then fed into Tencent’s deep neural networks to train a speech synthesis system (Text-To-Speech system).

At training time, the input sequences are real waveforms recorded from human speakers.

After training, we can sample the network to generate synthetic utterances.

At each step during sampling, a value is drawn from the probability distribution computed by the network.

This value is then fed back into the input and a new prediction for the next step is made.

Building up samples one step at a time like this is computationally expensive, but we have found it essential for generating complex, realistic-sounding audio.

Statistics from Unicef mentioned that there are almost 60 million left-behind children in China. That is almost the entire population of Italy.

Left-behind children are children who remain in rural China while their parents leave for work in urban areas.

They mostly see their parents once a year or every few years. The toll that this takes on both the parents and they are both physical and emotional.

WeBank, a bank that that believes in the power of innovation recognizes this societal problem and decides to use its innovative and financial capabilities to help parents and children bridge their emotional and physical gap.