Watson-Node-Red-examples

A collection of examples on how to use the Watson nodes in Node-RED.

These Watson nodes are used:

• Language Identification
• Language Translation
• Personality insights
• Speech to Text
• Text to Speech
• Tradeoff Analytics
• Visual Recognition

First you need an instance of Node-RED with Watson services. Therefore you need a Bluemix account. You can get that here: http://www.bluemix.net.

Once you are in Bluemix, go to Catalog and then go to the boilerplate section and select Node-RED Starter, in the next screen, give your application a name and click create.

You need to add the services you want to use to your application. Go to the dashboard and click on your application, then click on Add a Service or API. In the following screen select the service you want to use, and finally click on use. Wait for a moment to restart the application. When the application is started you can click on the URL to open Node-RED.

Another way of using Node-RED is installing it locally, which can be done by following this:

Check out http://nodered.org/docs/getting-started/ for full instructions on getting started.

sudo npm install -g node-red
node-red
Open http://localhost:1880

You have to install the Watson Nodes, which is described here: http://flows.nodered.org/node/node-red-bluemix-nodes and you have to make the services available in Bluemix.

DESCRIBE THAT

Note that Node-RED in BlueMix will behave slightly differently than Node-RED stand-alone:

1. The sets of available nodes differ, BlueMix has extra nodes for DB access, but does not expose the File nodes.
2. Node-RED in bluemix stores its persistent data (flows, libraries, credentials) in the co-installed Cloudant database named nodered. When using a Cloudant node with Node-RED on BlueMix, the list of available instances is automatically listed.
3. Node-RED in BlueMix has built-in credential management, so you don't have to worry about exposing your services authentication data, they will be filled-in automatically from the sevices' credentialds defined for the application in BlueMix.
4. Additional nodes in Node-RED on BlueMix are installed through cf and a specific procedure since there is no direct access to the npm package manager.

Language Identification

The Language Identification enables you to quickly identify the language text is written in.

This service can identify many languages: Arabic; Chinese (Simplified); Chinese (Traditional); Cyrillic; Danish; Dutch; English; Farsi; Finnish; French; German; Greek; Hebrew; Hindi; Icelandic; Italian; Japanese; Korean; Norwegian (Bokmal); Norwegian (Nynorsk); Portuguese; Spanish; Swedish; Turkish; Urdu.

In this example some random text is injected, identified by the Watson Language Indentification service and put the result to the Debug tab. In the following screenshots you can see how the nodes are configured.

In the following screenshots you can see how the nodes are configured.

The inject node:

You can use any text for this.

The Language Identification node does not need any configuration

And this is the output:

You can also copy the code for the flow here and import it from clipboard into Node-RED:

Language Translation

The Language Translation service enables you to translate text from one language to another. These languages are supported:

• The News domain - targeted at news articles and transcripts, it translates English to and from French, Spanish, Portuguese or Arabic.
• The Conversational domain - targeted at conversational colloquialisms, it translates English to and from French, Spanish, Portuguese or Arabic.
• The Patent domain - targeted at technical and legal terminology, it translates Spanish, Portuguese, Chinese, or Korean to English.

In this example some random text (in English in this case) is injected, translated (to French) and put the result to the Debug tab. In the following screenshots you can see how the nodes are configured.

The inject node:

You can use any text for this. I put Node-RED in double quotes, otherwise it would be translated as well.

The translation node:

The output from the debug node:

You can also copy the code for the flow here and import it from clipboard into Node-RED:

....Work in progress...More to come in a moment...

#Personality Insights

The IBM Watson Personality Insights service uses linguistic analytics to infer cognitive and social characteristics, including Big Five, Values, and Needs, from communications that the user makes available, such as email, text messages, tweets, forum posts, and more, go to https://www.ibm.com/smarterplanet/us/en/ibmwatson/developercloud/personality-insights.html for a full description and demo

The objective is to send the Personality Insights node the same data as for the demo (https://watson-pi-demo.mybluemix.net/) and to show the JSON object and its values match those of the demo

Create a new application using the Node-RED starter boilerplate option in Bluemix

Click on the icon of the application

Pick “Add a Service or API” and click the Personality Insights node under Watson section. Make sure your Application is listed and click on Use. Click on Restage.

Drag an inject node to the palette.

Drag the personality insights node to the palette.

Drag a debug node to the palette.

Join them as shown below

Double-click the debug node and replace ‘payload’ to ‘insights’. Also change “to” to “debug tab and console”

Double-click the inject node and change Payload to string. Go to https://github.com/NodeREDWatson/Watson-Node-Red-Samples/blob/master/files/Personality%20Insights/Personal%20Insights%20input%20text.txt and open the file and pasts the text to the blank field below the Payload field (this is the same text that is used in the demo).

Open a command line on your laptop and login to Bluemix with your username/password. Navigate to the organisation and space to where your application is located.

Type cf logs app_name

where app_name is the name of your application. Return to the Node-RED palette.

Click the Deploy button

Click on the debug tab to the top right of the palette.

Click the button next to the inject node.

You will see an output in the debug tab starting with '{ [msg.insights] : object }' followed by the JSON object. NOTE : you will not be able to see the full object

Go back to the command prompt and see that the full object has been shown.

Compare the items in the JSON object to those of the demo for Message Insights - you can see that the values shown are the same (or very similar).

Text to Speech

Overview

The Watson text-To-Speech (TTS) service produces an audio file from literal text. The spoken text can be emitted with a choice of voices and languages.

Node-RED Watson TTS node

The Node-RED node provides a very easy wrapper node that takes a text string as input and produces a binary buffer holding the spoken text audio stream in .wav format. The selection of language and voice are made through the node's properties editor.

Basic TTS Flow

In this first exercise, we will show how to simply produce a .wav file from input text through a simple web page generated using a Node-RED flow.

The first part of the flow will take text input from a web invocation and return the spoken text .wav file:

1. Create a new flow, let's call it TTS Web
2. Add an node to collect the incoming speech request. Set the URL property of this node to /tts/sayit This URL will be exposed below our BlueMix main URL. When invoked with query parameters such as ?text_to_say=Hello, they will be added as properties on the msg.payload object.
3. Add a node to extract the query parameter msg.payload.text_to_say and set it as the msg.payload. We do this because the TTS node uses the text in the msg.payload as input.
4. Now add a node. This node will generate the binary wav stream content to the msg.speech property.

The properties of the TTS node will let you select the Language and Voice to use.

5. Add another node to extract the msg.speech and place it in msg.payload. We will also set the HTTP response headers by setting the msg.headers to the literal string value [{ 'Content-Type', 'audio/wav'}]. This is required in order to let browsers know that this is an audio file and not HTML.

6. Finally, add a node. This node will simply return what's in the payload to the HTTP response. The completed flow should look like:

The flow code for this is in TTS-Lab-Basic.

Now try the flow:

• Open a new tab or window in your browser, and direct it to /http://xxxx.mybluemix.net/tts/sayit?text_to_say=Hello
• This should prompt you to save a file. Depending on how your browser is configured, it may save it automatically or prompt for a name and location. In any case, store or rename it with the .wav extension on your local file system.
• Then locate that file from windows explorer and open it with Windows Media Player, turn your you should

TTS Flow - enhancements: Input Parameter Checking

This flow has a caveat, which is that the flow will fail when the text_to_say query parameter is not set.

So, we will introduce a switch node between the [get] and change nodes. This node will check if msg.payload.text_to_say is set, and otherwise divert to a template node that will simply set the payload to a error text. You'll notice that adding the second otherwise rule has created a second output handle for the switch node, we'll connect this to a template node and then to the HTML Response node. The template node simply outputs a HTML message in a h1 header. Flow for this can be found in TTS-Lab-Basic The final flow will look like:

TTS Interactive Web UI

As an extension, we can build a flow that will present a dialog to the user with a prompt to enter the text to say, and return a HTML page with an tag which will play the generated audio.

For this, the basic flow which converts a text into speech audio wav file can be leveraged, and complemented with a HTTP web interaction. This is depicted in the flow as below::

We added a new HTTP input node, listening on the /talk URL, and modified the text-to-wav HTTP URL to /talk/sayit so that it doesn't conflict with the previous Lab. The choice node checks for the text_to_say query parameter, and when not present outputs a simple web page using the GetTextToSay template:

    <h1>Enter text to Say</h1>
       <form action="{{req._parsedUrl.pathname}}" method="get">
          <input type="text" name="text_to_say" id="" value="{{payload.text_to_say}}" />
          <input type="submit" value="Say it!"/>
       </form>

When a text_to_say query parameter is set, we generate an HTML page with a <audio> tag that refers to the /talk/sayit URL to get the audio wav file:

    <h1>You want to say</h1>
    <p><q>{{payload}}</q></p>
    <p>Hear it:</p>
    <audio controls autoplay>
      <source src="{{req._parsedUrl.pathname}}/sayit?text_to_say={{payload}}" type="audio/wav">
        Your browser does not support the audio element.
    </audio>
    <form action="{{req._parsedUrl.pathname}}">
        <input type="text" name="text_to_say" id="" value="{{payload}}" />
        <input type="submit" value="Try Again" />
    </form>

The complete flow is available at TTS-Lab-WebPage.

To run it, point your browser to /http://xxxx.mybluemix.net/tts/talk and enter some text.

Watson Visual Recognition

Overview

The Watson Visual Recognition service allows to analyse the contents of an image and produce a series of text classifiers with a confidence index.

Node-RED Watson Visual Recognition node

The Node-RED node provides a very easy wrapper node that takes an image URL or binary stream as input, and produces a set of image labels as output.

Basic Watson Visual Recognition Flow

In this exercise, we will show how to simply generate the labels from an image URL.

The flow will present a simple Web page with a text field where to input the image's URL, then submit it to Watson Visual Recognition, and output the labels that have been found on the reply Web page. The nodes required to build this flow are:

• A node, configured with a /reco URL
• A node which will test for the presence of the imageurl query parameter:
• A first node, configured to output an HTML input field and suggest a few selected images taken from the main Watson Visual Recognition demo web page:

    <h1>Welcome to the Watson Visual Recognition Demo on Node-RED</h1>
    <h2>Select an image URL</h2>
    <form  action="{{req._parsedUrl.pathname}}">
        <img src="http://visual-recognition-demo.mybluemix.net/images/horses.jpg" height='100'/>
        <img src="http://visual-recognition-demo.mybluemix.net/images/73388.jpg" height='100'/>
        <img src="http://visual-recognition-demo.mybluemix.net/images/26537.jpg" height='100'/>
        <img src="http://visual-recognition-demo.mybluemix.net/images/4068.jpg" height='100'/>
        <br/>Copy above image location URL or enter any image URL:<br/>
        <input type="text" name="imageurl"/>
        <input type="submit" value="Analyze"/>
    </form>

• The node, preceded by a node to extract the imageurl query parameter from the web request and assign it to the payload to be provided as input to the Visual Recognition node:

• And a final node linked to the output node. The template will format the output returned from the Visual Recognition node into an HTML table for easier reading:

    <h1>Visual Recognition</h1>
    <p>Analyzed image: {{payload}}<br/><img src="{{payload}}" height='100'/></p>
    <table border='1'>
        <thead><tr><th>Name</th><th>Score</th></tr></thead>
        {{#labels}}
          <tr><td><b>{{label_name}}</b></td><td><i>{{label_score}}</i></td></tr>
        {{/labels}}
    </table>
    <form  action="{{req._parsedUrl.pathname}}">
        <input type="submit" value="Try again"/>
    </form>

To run the web page, point your browser to /http://xxxx.mybluemix.net/reco and enter the URL of some image. The URL of the listed images can be copied to clipboard and pasted into the text field.

The complete flow is available at Reco-Lab-WebPage.