I live in Poland, where Google Assistant isn't allowed (operated) and I would like to write a simple script, which allow me to say: "Ok, google, muzyka stop". Next device will stop music play. Is it possible to write this in dialogflow or other system?
When creating apps for the Google Assistant I'd recommend staying away from use cases already covered by the Google Assistant. In this case controlling music is already covered by the Google Assistant and will already work with most music streaming providers.
If you'd like to add the Google Assistant to your device and add the option to control your device, a conversational building tool like Dialogflow is probably not the best tool. You may want to consider adding the Google Assistant and voice controls to you device use the Google Assistant SDK.
Related
I'm working on my own voice assistant based on the Google Assistant SDK. Now I'm wondering if it's possible to also communicate with custom Google Actions. (such as games and other actions that are not built-in to the SDK itself)
Thanks in advance!
Yes, when using the Assistant SDK, you have access to a fully functioning Assistant - including Actions.
Is that possible to connect google assistant with chromecast like netflix.
suppose i have already developed an app for google assistant, app name is test and suppose i have something in the test app to show like report.
Is this possible that,I would say hey google, show report then. It should automatically search chromecast devices nearby and connect to near by chromecast device then show the report on that connected chromecast device.
Yes, this is possible. Since August 2018, voice commands officially work with the following apps:
Netflix
CW
CBS All Access
HBO Now
YouTube
YouTube TV
Viki
Crackle
Red Bull
Starz
Google Play Movies & TV (for videos you’ve already rented or purchased)
To use voice commands with Chromecast, you need a device with Google Assistant (such as the Google Home or Google Home Mini smart speaker), plus a phone or tablet with the Google Assistant app installed. Voice commands also work with some smart TVs that have Chromecast built in (see Google’s website for a list of supported sets). Check this article for more details.
While casting from other sources works, an action is not able to provide content that can be casted to other devices.
After deleting all activity associated with the Google account linked to the AIY Voice Kit Pi project, the Assistant no longer responds to voice input queries. Now the Assistant responds, "Actually there are some basic settings that need your permission first. Just go into the Google Assistant settings on your phone for more details." I did not use a phone for initial setup of the AIY Voice Kit project. Why would I need to use a phone now? Is there a workaround that I can do using the web interface of my Google account?
You should be able to follow these instructions through the web to enable certain permissions in order to use the Assistant.
Open the Activity Controls page
Ensure the following toggles are on:
Web & App Activity
In addition, be sure to select the Include Chrome browsing history and activity from websites and apps that use Google services checkbox.
Device Information
Voice & Audio Activity
If you are using a corporate managed Google account with the Google Assistant SDK ask your administrator to turn on Web & App Activity (for everyone or just your account). Ours was turned off for everyone which prevents an individual user from selecting the "Include Chrome browsing history and activity from websites and apps that use Google services" checkbox.
Google Admin G-Suite
I'm playing around with assistant SDK 0.1.0 .. Are "Actions on google" app supported on it? When I try to speak to my app it tells me " isn't supported on this device"
Yes Actions on Google are supported on the Assistant SDK. Saying things like "talk to Number Genie" will hand off the conversation to a third party app.
Are you sure the device is transcribing the action's name correctly? It should behave like any other surface.
By continuously I mean something like continuous.c, the default program of pocketsphinx. As I tried google voice recognizer or other apps like cortana I know the user should press the record button, then the voice recognizer records his sound for seconds, then sends it to server and so on. But in pocketsphinx, when you run the program, it's listening to you until you terminate it.
I know the pocketsphinx works offline and google and cortana work online. So I am interested to know, is it possible to do continuous voice recognition with online api's like what pocketsphinx does in offline, or not?
If I want to build a robot that works with voice commands, is it possible to use google asr api considering that there is no record or any button to start recording voice commands for the robot(like what we do in smartphones or computer)