With network connection
get assist in a blink



unique users per week



Assist you
in your language


How the actions works
It is a new type of application with a voice-based design interface that interfaces with Google Assistant. After the user indicates to Google Assistant that they want to use a specific Action, Google will search the Actions On Google platform to see if there is an Action with the corresponding name. The user is then directed by the Google Assistant to the Action interface. From this moment on, the role of the Google Assistant has shifted to assisting with speech recognition and information delivery. The task of recognizing the user's input intent and giving the corresponding response is transferred to the Action designed by the developer.
On devices that support Google Assistant,
Natively built in an officially supported third-party platform known as Actions On Google (AoG).

As a user requests to interact with a action on this third-party platform, the system will automatically search for the corresponding action to be serived. Once the platfoem found, the control of the response will be handed over to the third-party action. In this point, the Google Assistant's role has shifted to an intermediary that performs speech recognition and presents third-party action content.

In practice, the entire interaction process is information transfer through API in JSON format.
As the user interact with the action, the following pipeline with be practice
Until the request of the user is fulfilled or the action is interrupted, each round of the conversation is perform in this way:
  • 1. User asked to the Assistant
  • 2. Assistant transfer the (recognized) sentences send to the Action
  • 3. In the structure of the action, we use DialogFlow which containing the pretrained modle to judged the intent of the user
  • 4. To perform the complex responds, the DialogFlow narrow down the sentence and send the coresponding parameters to the Fulfillment where will perform data fetching or API call
  • 5. Fulfillment return the relative Intent responds to the Assistant
  • 6. Assistant recived the responds and show it to the user
Based on requirements, there are 4 types of the structure are constructed.
1. Without the requirement of the fetching data , which is the simplest structure.
2. With the requirement of the fetching data , it fetch data processing by APIs in the same Cloud Function.
3. Perform the seperation of frontend and backend, the service of fetching data is now packged in one independent Cloud Function, the other one is focusing on the DialogFlow logic and responds.
4. In this structure, the Cloud Function fetch data automatically in preset time and perform the data update to the Firebase Realtime Database, the other one is focusing on the DialogFlow logic and fetching the data from the database.

Google Developer Group DevFest

Host by GDG Taiwan

In this online speech:
Share the brief process of VUI design and how to apply it to develop a Google Assistant voice app.
At the same time, I share the info of complete tutorial for the listener.
Onsite page

In this online speech:
Share my experience in developing Google Assistant conversational Actions.
In the same time ,I take the audience to create a simple voice application as well.
Onsite page

Publicized by Google Taiwan

* Actions with the mark are available in Chinese(Taiwan) only

July 31, 2019

GOOGLE Assistant connects local services in Taiwan,if you need some fun just ask
(Google 助理連結台灣在地服務 吃喝玩樂一問就知)

 • Meal Decider
 • Cows and Bulls
 • Master of riddle *
 • Sport Meeting of Brain *

Check it out

March 31, 2020

Protecting health together, making good use of GOOGLE to make life not boring anymore
(共同守護健康 善用 Google 不無聊)

 • Text Solitaire *

Check it out

Action library

They can automatically change language by user preference!

User feedback

From users around the world