Hi all,

first of all great labs!

One thing I came across: The visual recognition lab would needs some updates

• The linked pictures are not available any more in the Get Image URL Template Node
• Template screens in Node-Red look a bit different now (Syntax Highlight, Format)

The Extract and GenerateReply in the Text to Speech flow has a bug. The payload is an object that carries both filename and the text to process. So in the Extract leave the text where it is. Then the GenerateReply output should be referring to the {{payload.text_to_say}}

You want to say

<p><q>{{payload.text_to_say}}</q></p>
<p>Hear it:</p>
<audio controls autoplay>
  <source src="{{req._parsedUrl.pathname}}/sayit?text_to_say={{payload.text_to_say}}" 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.text_to_say}}" />
    <input type="submit" value="Try Again" />
</form>

Are websockets mandatory?

Is an Express application required?

Thanks!

If you set the audio file name, then the extension / mime type will get recognised, and the user will only be given a selection of audio players to use. I suggest using a function with the following code -

msg.headers = {
'Content-Type' : 'audio/wav',
"Content-Disposition" : "attachment;filename=ttsaudio.wav"
};
return msg;

If you want to try to play the file in the browser (but that will need a lot more client side html & javascript) try
"Content-Disposition" : "inline"

Highlight Button
You will need a Classifier ID, this can be obtained by clicking the Classifiers button and the Classifier ID is shown. Circle the button to highlight.

Minor Typo in heading
"Connecting to a existing NLC Service on Bluemix" should read
"Connecting to an existing NLC Service on Bluemix"

Formatting suggested change from
"In this lab we will assume (for now) that you have created a NLC Service in Bluemix and now wish to (a) check it's status and (b) ask it a question via calls in Node-RED. "
to
"In this lab we will assume, for now, that you have created a NLC Service in Bluemix and now wish to:

• - (a) check it's status and
• - (b) ask it a question via calls in Node-RED"

Typo
"Open your Node-RED flow editor and drag/drop an two Inject nodes, two Function nodes, one http request node and one Debug node and join up as shown below :"
should read
"Open your Node-RED flow editor, then drag/drop two Inject nodes, two Function nodes, one http request node and one Debug node and join up as shown below :"

No blank
Double-click the top Inject node and select Blank from the option - I do not see blank in the options and therefore I have chosen string with nothing.

get NLC status
missing "; from the end of the line
also you should use your own classifier id

second function button
The end of the line is not shown on screen please ensure that you copy to end of line
also you should use your own classifier id

The project has things like node-RED_labs where you are using UPPERCASE, underscore and dashes at the same time.

I recommend you to use snake case so it should be: node_red_labs

Minor point but I would highlight the region setting.
Maybe highlight the whole line
Note For this exercise please check that your Bluemix region is set to US South

Image of alchmey feature extraction node is out date. Please change fe_node.png

The http request part of the lab is broken. You can no longer add an inject node to the http request. Please fix lab asap !

Steps missing to add TTS to Node Red app so users are asked to enter username and password

https://github.com/watson-developer-cloud/node-red-labs/tree/master/basic_examples/alchemy_api_image_analysis Last link to get the complete flow gives 404. The correct link should be all lower case letters in the name of the flow.json ..

Another issue I have seen is that within the Speech-to-Text lab the resulting transcription is:
"the space shuttle enterprise was the first orbiter of the space shuttle system". The .wav file
contains though a longer recording after a quick pause.

Not sure whether this is about the inactivity_timeout parameter or the recognition type used within the Node-Red node. But we should transcribe the whole .wav in our sample or change the sample ;-).
Best though would be the ability to pass parameter to configure the transcribing (inactivity timeout, etc.). Perhaps this is already possible?

Thanks.

When I follow the steps in the 'text to speech' lab, the .wav file is created but I receive the following error when trying to play it with Windows Media Player:

Windows Media Player cannot play the file. The Player might not support the file type or might not support the codec that was used to compress the file.

Any suggestions appreciated.

I previously was able to easily connect my Visual Recognition node in Node-Red as I had added it as a service for my product. Now I am getting "could not authorize" the visualization node. After conferring with bluemix support back and forth, they told me it is because Node-Red updated but did not also update its support for the visual recognition node. This is really unfortunate as my entire tutorial which involved zero coding in Node-Red is now useless in demonstrating how easy it was!!

I created an app from the Bluemix Node-RED Starter Boilerplate. In the Node-RED editor I don't see Watson nodes such as AlchemyData News nor Dialog.

I want to use in the STT example my microphone instead of the a .wav file as you described it. What change needs to be done in the Example

Hi,

May I please see a sample flow in node red with the following watson apis?

• natural language classifier
• dialog
• text to speech
• speech to text

Cheers,

Perhaps the lab instructions could be updated to say to edit the image analysis node to set Detect = Faces as by default this field is blank.

The links to the example images in the visual recognition lab aren't working. It doesn't really affect the lab as an external url can be used but perhaps these could be corrected or removed?

http://visual-recognition-demo.mybluemix.net/images/horses.jpg
http://visual-recognition-demo.mybluemix.net/images/73388.jpg
http://visual-recognition-demo.mybluemix.net/images/26537.jpg
http://visual-recognition-demo.mybluemix.net/images/4068.jpg

Hi,
The documentation doesn't show how to set msg.alchemy_options for Alchemy vision.
Could you please provide an example how imagePostMode can be set as child of msg.alchemy_options?
Thank you!

Add a CONTRIBUTING.md file telling users how they can contribute to the project, the code style, and tests if need.

See an example:

https://github.com/watson-developer-cloud/nodejs-wrapper/blob/master/CONTRIBUTING.md

Hello,

On the basic labs page, the 'personality insights' link return an 404 error code with the message 'can't find the page'. The 'visual recognition' link also seem to be broken, if you wouldn't mind taking a look at that please too?

Thanks

Re: issue #69, there are still two dangling links that I could not relate to anything in the repo:
basic_examples\relationship_extraction\README.md re_input_file
basic_examples\relationship_extraction\README.md images\fe_inject_blank.png

Create a new Application using the IOT Starter Boilerplate.
Why?? Surely you can use the existing boilerplate. Then you already have the NLC bound!! You can then start at "Drag a Natural Language Classifier (NLC) node to the palette"

The instructions ask the http request to be set to /tts/sayit and then in the invoke instructions it states /talk/sayit... need to keep it either 'tts' or 'talk'

Broken link near the end of the Tradeoff Analytics sample

Link text - Watson tradeoff
url - https://www.ibm.com/smarterplanet/us/en/ibmwatson/developercloud/doc/tradeoff_analytics/inout.shtml

"A completed flow file card be found here". Is the final line but the url for here leads to 404

Hans - can you add a full flow file for /watson_advanced_labs/alchemy_output_table so people can cut/paste it to their node-RED installs ?

Change
"Add 3 more injectm NLC and Debug nodes as shown below"
to
"Add 4 Inject nodes, 3 Classifier nodes and 3 debug nodes as shown below:"

Alter text
"Change the second NLC node to contain List in the dropdown"
to
"Change the first new Classifier node to List in the dropdown"

Alter text
"Change the next NLC to Classify and the final one to Remove. "
to
"Change the next classifier node to Classify and the final one to Remove. "

Alter text
"NOTE : For the Remove Inject node you must change the Inject string to a Classifier that exists (you noted one down earlier)." add highlighting to draw attention to the classifier id.

1. https://github.com/watson-developer-cloud/node-red-labs/tree/master/advanced_examples/tradeoff_analytics_widget it would be good to have the script of the Set Problem stored in a file to be readily downloaded and used by participants
2. https://github.com/watson-developer-cloud/node-red-labs/blob/master/advanced_examples/alchemy_output_table/README.md search for the text "The HTML code can be downloaded here and looks like this:" the here link is broken.

The base documentation for Visual Recognition V3 is missing. Also the flow in the node-red-bluemix-starter set hangs on the node-red node an example included have tried it on four of the examples.

Add instructions in the Node-RED intro lab about how to secure the node-red URL in BMix

During testing of more advanced labs we came across the message "Alchemy API request error: daily-transaction-limit-exceeded" when sending data to the AlchemyAPI Node-RED node. This is important if end users of the labs reach this limit. At present we do not know what the actual limit is and whether Bluenmix is enforcing it or AlchempyAPI team (indirectly).

You need to be more specific in the instructions. I have assumed that you meant for me go back up the levels until I got a Download Zip button!

Is there another method?

I want to change my index.html page to display my Node Red flow. How do I add these two flows?

1.) Website Flow:
HTTPRequest->Template->HTTPResponse sequence

2.) NLC Flow:
[{"id":"b82d2276.47d2e","type":"inject","name":"","topic":"","payload":"","payloadType":"none","repeat":"","crontab":"","once":false,"x":409,"y":468,"z":"7ada38b4.8525c8","wires":[["e211d200.1dee3"]]},{"id":"a565f810.5a9a08","type":"inject","name":"Ask","topic":"","payload":"Is it hot ?","payloadType":"string","repeat":"","crontab":"","once":false,"x":409,"y":526,"z":"7ada38b4.8525c8","wires":[["f6ceadee.09315"]]},{"id":"e211d200.1dee3","type":"function","name":"get NLC status","func":"msg.url="https://gateway.watsonplatform.net/natural-language-classifier/api/v1/classifiers/D385B2-nlc-530";\nreturn msg;","outputs":1,"noerr":0,"x":598,"y":467,"z":"7ada38b4.8525c8","wires":[["9eb5ad14.614a5"]]},{"id":"f6ceadee.09315","type":"function","name":"Ask NLC question","func":"msg.url="https://gateway.watsonplatform.net/natural-language-classifier/api/v1/classifiers/D385B2-nlc-530/classify?text=" + encodeURI(msg.payload);\nreturn msg;","outputs":1,"noerr":0,"x":585,"y":525,"z":"7ada38b4.8525c8","wires":[["9eb5ad14.614a5"]]},{"id":"9eb5ad14.614a5","type":"http request","name":"","method":"GET","ret":"txt","url":"","x":811,"y":489,"z":"7ada38b4.8525c8","wires":[["f3ac9f51.0c536"]]},{"id":"f3ac9f51.0c536","type":"debug","name":"","active":true,"console":"false","complete":"false","x":1020,"y":491,"z":"7ada38b4.8525c8","wires":[]}]

Reference: https://github.com/watson-developer-cloud/node-red-labs/tree/master/basic_examples/natural_language_classifier

I have tried the samples with both .flac & .ogg formats but the results are always .wav.
I changed the option in the node

and the type in "set headers"

but the output was still a .wav.

A sample flow is required for the new conversation node.

Hi guys. I have followed the Visual Recognition tutrial however, the sample HTML is not producing the label_name or label_score variables

Suggestions?

Natural Language Classifier is a GA service since Aug 4. The Beta service was removed on Sept 4. However the example here uses the Beta service

Section on Watson contribution nodes needs to be amended to point at our new repository. Plus add in mention for box and Dropbox nodes.

Hello,

On the 'Watson Node-RED services labs' page, the links 'Tradeoff Analytics', 'Text to speech', 'Visual recognition', 'Alchemy Vision', 'Alchemy Feature Extraction', 'Natural Language Classifier BETA service' and 'Relationship Extraction' don't seem to be working correctly.

The dropdown for inject node no longer has a blank option

I had to give my app a name before I could create the app and continue. I assume there is no restriction on naming

Since the folder are within a NODERedWatson organization. You don't need to use the word watson in each folder. In fact the examples are using the Watson Developer Cloud (which include Alchemyapi) not Watson in general (WEA, WDA, etc..).
I would recommend to use have the next layout:

/README.md        # Overview, Getting Started, how to run the examples, license, etc...
/images           # folder containing the images(snake_case)
/advance_examples # instead of watson_advanced_labs
/basic_examples   # instead of watson_services_labs
/LICENSE          # no need to use .txt

There is no screenshot that highlights the expected lab outcome

Once deployed to Bluemix, how can I keep this app updated? Would a redeploy cause the new flows I may have added to disappear?

https://github.com/watson-developer-cloud/node-red-bluemix-starter

Hello,

On the Introduction to Node Red page the link on line 72 'Check out this page' returns a 404 error 'File not found'.

When looking at the different example folders we should try to call the example instructions README.md since it will automatically loaded/parsed by Github. When using lab_tradeoff_analytics_widget.md users need on it click render it.