🗄️Ez Google AI

🔗Links

Introduction

There has never been a more exciting time in technology. We are experiencing one of the most significant shifts in history, where AI is creating entirely new ways to solve problems, engage customers, and work more efficiently. Google Cloud is ready to help organizations build a new way forward in an increasingly AI-driven world. We are glad to bring Google AI technology to the Bubble community. We decided that to have all tools in one place is best practice, so Ez Google AI combines several actions in one plugin:

Document AI
Speech To Text
Get Text from Image
Get Safe Search
Get Logo Search
Get Landmark Detection
Create Cloud Storage
Upload to Cloud Storage

You enter your private Key, Client ID, and Project ID in one place and it is ready to use Ez Google AI.

Instruction and Description

Create an account on Google Cloud. You can read more about the terms and conditions of registration, and further management of accounts, at the following link.
Enable APIs that You need to use (Image AI, Text To Speech, Speech to Text, Document AI, Cloud Storage).
Create Processor from Cloud Console(Needed for Document AI).
Create a Service Account. - Add Credentials to IAM.
Enable billing (needed, but You get from Google trial period and free calls).
Download a private key as JSONHints for Speech to Text.

Auto Authentication

Open the private key JSON file with a text editor and copy/paste the following parameters from your file to the Plugin settings: - client_email - project_id - private_key,

Including the -----BEGIN PRIVATE KEY-----\n prefix and \n-----END PRIVATE KEY-----\n. How it’s WorkEz Google AI contains Actions, Specify trigger (button, event, etc.), and starts Plugin Action.

Instruction to Document AI

Specify element Action from WorkFlow, add Action “Document AI- Get Text From Document”

Now You need to fill in the fields “Processor ID” and “Link to File” and choose “Location”

Processor ID field - Here you need to paste Your processor ID from Google Cloud Console, “My Processors” tab. If You don’t have a Processor ID, create one with the “Document OCR” type.
Location - drop-down. By default, Document AI stores and processes resources in a US location. If you choose the European Union location, Google will store your data and process it only in the European Union.
Link to File - Bubble Uri to document Example: “//0341b7fbc64b39338695c11fb9a66136.cdn.bubble.io/f1697709115621x854084300796481200/How%20much%20is%20the%20fish.pdf”

After the plugin ends execution, you will get the response key-value object with the next keys: Text, Language code, and error.

Text - Contains all found text in one string
Language Code - Returns the language code of a document
error - Returns “null” for no error, or a String with an error definition

Instructions to Image AI

Specify element Action from WorkFlow, add Action “ImageAI- Get Lable Search”, practically all Image AI actions have the same Input Fields.

Now You need to specify input fields “Dynamic Image”, “Max Result”

Dynamic Image - Link to bubble Uri(ex. “//0341b7fbc64b39338695c11fb9a66136.cdn.bubble.io/f1697718926510x406478102390087000/items_test.jpg”)
Max Result - Specify How many objects Image AI must recognize.

After the plugin ends execution, you will get the response key-value object with the next keys: Found Items and error.

Found Items - Return the string of Objects found on the image.
error - Returns “null” for no error, or a String with error definition.

Instructions for Cloud Storage

Specify element Action from WorkFlow, add Action “STORAGE- Create Storage”.

Now You need to specify the input field “Storage Name”

After the plugin ends execution, you will get a response key-value object with the next keys: Status and error.

Status - Returns “yes/no” result if bucked created
error - Returns “null” for no error, or a String with an error definition

Upload Storage Action

For Uploading a file to Bucked You need to specify the Input Fields:

Link to File - Bubble Uri (ex.”//0341b7fbc64b39338695c11fb9a66136.cdn.bubble.io/f1697718926510x406478102390087000/items_test.jpg”)
File Name - Name of the uploaded file to the bucket.
Bucket - Name of Your bucket(Place, Where You want to store a file).

After the plugin ends execution, you will get the response key-value object with the next keys: gs Link, status, and error.

-gs Link - Google Storage Link(Needed for Speech to Text action) ex. “gs://{bucket name}/{filename}”.

-status - Returns “yes/no” result if bucked created.

-error - Returns “null” for no error, or a String with an error definition.

Delete Storage Action

For Deleting a file from Bucked You need to specify the Input Fields:

Bucket Name - Bucket from which You want to delete a file.
File Name - Name of File.
Mime - Specify file extension without “.” (ex. pdf, jpg, png, etc.).

After the plugin ends execution, you will get a response key-value object with the next keys: status and error.

status - Returns “yes/no” result if bucked created.
error - Returns “null” for no error, or a String with an error definition.

Instructions to Text to Speech

Specify element Action from WorkFlow, add Action “Text to Speech”.

Now You need to specify the following fields:

Home Url - For “cross-origin” must be bubble.io home URL (needed for creating a link to an audio file)
Text.
Language Code - Choose Your Voice https://cloud.google.com/text-to-speech/docs/voices.
Gender - Choose the Gender of the speaker

After the plugin ends execution, you will get the response key-value object with the next key.

file - Returns link with your Audio File (ex. “https://meta-l.cdn.bubble.io/f1698044891872x337292678477508540/EazyCode.mp3”)

Speech To Text Action

Now You need to specify the following fields:

gcsUri - gs Link to your Audio File, for now, google can ONLY encode audio files from Cloud Storage Bucket, so the link must look like this: “gs://cloud-samples-data/speech/brooklyn_bridge.raw”
Language Code - BCP-47 language code, e.g. en-US
Encoding - Google Speech-to-Text API can handle specific types of speech encoding.
Model - You must set the model field to one of the allowed values. https://cloud.google.com/speech-to-text/docs/reference/rest/v1p1beta1/RecognitionConfig
Profanity Filter - If enabled, Speech-to-Text will attempt to detect profane words and return only the first letter followed by asterisks in the transcript (for example, f***). If this field is set to false or not set, Speech-to-Text will not attempt to filter profanities.

Enable Automatic Punctuation

Transcript - Returns text string for Your Audio file.
Error - Returns “null” for no error, or a String with an error definition.

Hint for Speech to Text

Speech to Text works only with “gs” links. First, create a bucket or upload an audio file to storage, get “gs” link(ex. gs://bucket/filename). If enabled, Speech-to-Text will attempt to detect profane words and return only the first letter followed by asterisks in the transcript (for example, f***). If this field is set to false or not set, Speech-to-Text will not attempt to filter profanities.

Hint for Text to Speech

You can specify voices and Languages from https://cloud.google.com/text-to-speech/docs/voices You need to copy/paste the Language Code without “”.

Hint for Document AI

You need to create a Processor with the “Document OCR” option. Location is a spec for where Your Site will OCR document. Document AI supports the following file formats: