The Google speech recognition app that eliminates the wait for call centers
“Please wait so as not to lose the priority acquired”. One of the most hated phrases in the world, which everyone will have to listen to while waiting for an operator. Unnecessary waits that can sometimes last several minutes, if not hours. But dear Google has already thought of a solution with which he can come to our aid. This is a particular application of the speech recognition algorithm that will wait for us to answer the call center.
The new feature is called Hold for me , it was announced on the Made by Google Twitter profile a few days ago. The video shown shows how the mode can be activated during the call to the call center and allows it to be resumed as soon as the operator answers the call. This interesting feature will be available (at the moment) only on US-made 5G Google Pixel. It will therefore be necessary to wait a little longer to be able to have this functionality on your smartphone.
Hold for Me saves you, and your ears, hours of elevator music
How to activate Hold for me mode
Hold for me is activated directly during the call phase. Here are the steps to follow described in Google support.
- Enter silent mode in your smartphone
- Open the phone app.
- Dial the call center number.
- When you are put on hold press Hold for me -> Start.
- While waiting, the display will show "Do not hang up".
- When the operator is available for the call, the message "Someone is waiting to talk to you" will appear on the display
Google is also keen to clarify that no conversation will be saved in its databases and therefore all privacy regulations will be respected.
The Speech Recognition System
The technology developed by Google is part of an enhancement of the Speech Recognition algorithms, in this case used to spare us the wait with the call center. During the phone call, in fact, the human voice is recognized as such so it will no longer be necessary to wait. Let's see in more detail how this particular algorithm works.
An Automatic Speech Recognition (ASR) system is used to convert spoken words into text. Each ASR system consists of three processes:
- extraction of the characteristics of the voice being heard,
- training phase in which the system is “trained” to hear different types of voices
- correspondence between the expected output and the identified one.
To this system is added the peculiarity that allows the algorithm to recognize a human voice. This purpose can be achieved by using a neural network which is “taught” to distinguish the typical metallic voice of a robot from that of a human.
Other applications
The use of speech recognition to avoid waiting in call centers is just one of the latest applications of this type of algorithm. The system is in fact used in various fields.
In the education sector, for example, it can be used to teach students the correct pronunciation of a foreign word. Or yet another use may be to help those with physical disabilities use a speech synthesizer rather than a keyboard. We cannot fail to mention the possibility of using this algorithm in the field of robotics, for remote control. Furthermore, reading these lines will surely have come to mind the use of voice assistants such as Siri or Alexa. Well, even the latter exploit a technology of this type.
An interesting application can also be used to identify the symptoms of Covid-19 simply by analyzing the sound of the cough. We talked about it here .
Applications for this type of technology are increasingly growing. Over the years we will see more and more and more and more evolved.
The article The Google speech recognition app that eliminates the wait for call centers comes from TechCuE .