Do you plan on supporting the Dialogflow CX API?

Alternatively, is it easy enough to adapt this for CX?

Thanks for developing this!

Hello,

I am trying to integrate my dialogflow bot into our website by using one of your applications, but I am running into different problems.

• I am getting double responses.
• The bot’s logo is not showing up.
• Lastly it takes enormously long to get a response and sometimes doesn’t get any response.

I hope you can help me fix these problems or maybe you know what can cause these.

To be clear I tried the df-btn script, dialogflow-web-v2 and the source: https://jipsnel-assistant.ui.dialogflow.cloud.ushakov.co/
(It is a dutch bot tho, so that might give a problem for you to test it?)

Hello.
First, thank you for your awesome work, very useful.

When running the app, when using the mic to communicate, often when the input sentence is long, the Speech-to-Text stops in the middle of my sentence, when there are silence periods between words.

I was wondering if there is a parameter or config, that can make the STT ignore short silence periods and only end the transcription when the silence is longer than N seconds.

Thank you for your time,

Nelson

Hi
We are faced with the problem of limiting the length of the feedback speech. After pronouncing about 200 characters, the bot stops voicing. And then he does not voice the following response messages.

How we can fix it? Maybe there is some setup of speech lenth in project config?

People are requesting a feature to play Dialogflow's text-to-speech (instead of the browser one)

However:

• The implementation on Google's side seems incomplete. For example Webhook responses do get the audio, but platform-specific don't:

• Actions on Google don't work properly with Dialogflow and therefor do not get audio feedback as well:

So ladies and gentlemen, what are your thoughts on this? Thanks

How to add a gateway for Dialogflow? I'm unable to find the gateway key.

npm WARN deprecated [email protected]: request has been deprecated, see request/request#3142
npm WARN deprecated @hapi/[email protected]: joi is leaving the @Hapi organization and moving back to 'joi' (hapijs/joi#2411)
npm WARN deprecated [email protected]: Chokidar 2 will break on node v14+. Upgrade to chokidar 3 with 15x less dependencies.
npm WARN deprecated @hapi/[email protected]: This version has been deprecated and is no longer supported or maintained
npm WARN deprecated @hapi/[email protected]: This version has been deprecated and is no longer supported or maintained
npm WARN deprecated @hapi/[email protected]: This version has been deprecated and is no longer supported or maintained
npm WARN deprecated @hapi/[email protected]: This version has been deprecated and is no longer supported or maintained
npm WARN deprecated [email protected]: fsevents 1 will break on node v14+ and could be using insecure binaries. Upgrade to fsevents 2.
npm WARN deprecated [email protected]: https://github.com/lydell/resolve-url#deprecated
npm WARN deprecated [email protected]: Please see https://github.com/lydell/urix#deprecated
npm WARN deprecated [email protected]: core-js@<3 is no longer maintained and not recommended for usage due to the number of issues. Please, upgrade your dependencies to the actual version of core-js@3.

Hi @mishushakov ,

First thing first, This is an awesome project, thank you.

Now for the issue, I have create a DialogFlow Agent to support the knowledge base provided by dialog flow.

I followed the steps mentioned and it does respond back to the small talk options but knowledge base its not responding to the question.

I was wondering if that is supported by this project or is there something I am missing

https://cloud.google.com/dialogflow/docs/knowledge-connectors

Regards,
Wenceslaus Dsilva

I select an image for be the avatar for my agent but this image not showing.

Using the web demo link the image is showing like this:

the image in only a .png and i tried to change with another file but i had no success

edit: I changed the image several times with the same image I had chosen. In one of those times it somehow worked.

Hi

How can we change the voice of the app?

Can it use the standart voice, that was choosen from the list in dialogflow agent settings?

Where do you find the gateway URL? I've primarily been testing via Slack and WhatsApp so this is my first attempt at using a chat UI like this. In my console I can see my project ID, my API key, etc, but no clue where I would find the gateway URL.

Sorry I did try and rtfm via the docs on Dialogflow but no help.

Where and how to integrate "Dialogflow Gateway", there is not proper instructions on it.
The Dialogflow gateway link in you repo, i.e https://dialogflow-gateway.ushakov.co/ is not working.

the fist step in your application is to change the gateway in config file, but how to create a gateway. Please say the proper way to generate like gateway URL link..??

Thanks in advance.

Hey, I would like to change the UI of the gateway: the colours and background. How can I do it or where di I update the CSS for it?
Also, I would like to disable the voice output too.

Would you be able to have a quick look at https://crypto-haven-248819.ui.dialogflow.cloud.ushakov.co & see why it's not working. I get the message "Agent for project 'crypto-haven-248819' not found." but I'm pretty sure ive linked up an agent correctly.

I am getting this error for any command like

dialogflow-web-v2@ serve /app
vue-cli-service serve --open

sh: 1: vue-cli-service: not found
npm ERR! code ELIFECYCLE
npm ERR! syscall spawn
npm ERR! file sh
npm ERR! errno ENOENT
npm ERR! dialogflow-web-v2@ serve: vue-cli-service serve --open
npm ERR! spawn ENOENT
npm ERR!
npm ERR! Failed at the dialogflow-web-v2@ serve script.
npm ERR! This is probably not a problem with npm. There is likely additional logging output above.

Hey @mishushakov , I noticed that the chat input form can still send inputs although suggestion buttons pop up. This might break the flow as the user is free of inputting whatever he wants and might not press any suggestion. Is there a way I can implement a form freezing up to the point the user presses the suggestion?

To fix #30 allow to record audio and send it to Dialogflow 👌

We're using the embed code on our site and we've noticed that the mic input feature is not working on embed. The mic does work on the bot website source.

See issue: https://issuetracker.google.com/issues/165312268

Hi, not sure if its just me or not but im always getting the Reply as a double up.
Ive taken a look, but cant quite figure out why it would be doing this.
In google assitant simulator, or Dialogflow i only get the single reply.

Cheers

Installed dialogflow-web-v2 with the instructions given and it is successfully runs on development.
But voice works with firefox and not with chrome.

How do you change the default language, when accessing the bot? It looks like, that the default language is "en". But we want to change it to "de"

Hello there. ✋

First of all, thank you very much for this great project. We have been testing it in a project and it's perfect.

I wonder how can we use Lists responses.

In order to use cards, we do:

const { Card, Suggestion } = require("dialogflow-fulfillment");

However, Lists are not part of dialogflow-fulfillment and I didn't find any example... only the GIF at the readme.

Error: [BABEL] C:\Users\athen\Datascience\Dialogflow Bot\New folder\dialogflow-web-v2-master\src\main.js: Package exports for 'C:\Users\athen\Datascience\Dialogflow Bot\New folder\dialogflow-web-v2-master\node_modules@babel\helper-compilation-targets' do not define a '.' subpath (While processing: "C:\Users\athen\Datascience\Dialogflow Bot\New folder\dialogflow-web-v2-master\node_modules\@VUE\babel-preset-app\index.js")

How to fix this?

Hi Mikhail,
Thank you for coming up with this library to support v2 apis.
I am trying this with a sample dialogflow agent that uses dialogflow-fulfillment-nodejs library as webhook.
I have some intent handlers that return a payload for listSelect for ACTIONS_ON_GOOGLE platform. The sample code looks like below:

agent.add(new Payload(agent.ACTIONS_ON_GOOGLE, {
        "expectUserResponse": true,
        "richResponse": {
          "items": [{ "simpleResponse": { "textToSpeech": "Please select an option from below list." } }]
        },
        "systemIntent": {
          "intent": "actions.intent.OPTION",
          "data": {
            "@type": "type.googleapis.com/google.actions.v2.OptionValueSpec",
            "listSelect": {
              "items": [{
                "optionInfo": { "key": "Option1"},
                "title": "Option1"
              },
              {
                "optionInfo": { "key": "Option2" },
                "title": "Option2"
              },
              {
                "optionInfo": { "key": "Option3" },
                "title": "Option3"
              }]
           }
        }
     }
}));

However, an empty text response is rendered instead of list.
The fulfillment[component].platform in the response in index.js of dialogflow-gateway-master is returned as PLATFORM_UNSPECIFIED.
What should be the platform and format of payload in order to render the list correctly?

Dialogflow v2 API has changed how the media is returned, see: https://cloud.google.com/dialogflow/docs/reference/rest/v2/projects.agent.intents#responsemediaobject

web app voice recognition is just working in google chrome. What should do for that?

Should the Voice Recognition be working? I tap the microphone, but when I speak nothing happens.

Thank you!

Hey,
some people reported, that their app suddenly broke
i proposed and shipped an update months ago already
but some are still left behind

i would use the opportunity to remind you, that i offer free, hosted integration that is updated automatically, here

to update your self-hosted integration, run git pull
i recommend you watching this repository, so you are in check with any updates or issues. You can do so by pressing the "Watch" button on top of the page

Let me know if you need any assistance. Thanks

Hi, is there any way to send a message to dialogflow from the users side automatically when the user opens the page, for example, when the user opens the page, he automatically says hi to dialogflow

Hi! first of all thank you so much for this project, looks really promising!

I have a question I couldn't find anywhere, I want to run the both in my app, where the user is already logged in, so I would like to get its user id in all the webhooks calls so I know how to filter them,

is there any way to achieve that?

Thank you!

"

The deployment was not found" appears when I try to run the Web using run serve. I'm using my own Dialogflow agent for the gateway.

Is there any way to get a date input from the calendar as Carousel Selector ?

Hi,

Thanks for the great work.
I was wondering if there's any possibility to access the user's ID or Userstorage by this framework?

Thanks in advance,

Bas

Hi there,
I'm currently been experimenting on dialogflow web v2 agent. So far, I couldn't manage to send back an audio media content attachment as a message response to the web agent via Dialogflow's fullfulment index.js

I've tested it on my google assistant running on my google pixel and it worked so I was wondering if you could share with me how to integrate it with the web agent?

Thank you!

IAM permission 'dialogflow.agents.get' on 'projects/small-talk-hqbbmy' denied.

Following error when accessing from localhost

Does it support account linking?

Make it possible to queue tracks like in Google Assistant

How to fetch the conversation from dialogflow to here, i.e the intents created and fullfilments from dialogflow to this project.
Also is there any way to call REST API / Webhooks here??

Hi Mishushakov,
thank you for your great work. I just started playing around with your web interface and I really like it. I'm just wondering if you can commit a native version of your work. As far as I've read, by using Vuejs Native script library it involves a couple of changes in a couple of lines to turn your web interface into an android or iOS app. But when I try, I face a package conflict.
This is not an issue but rather a suggestion. If you could add this option would be amazing.

There is a problem when using iframe.
When sending a message or selecting it, the parent window will also scroll synchronously.

Is this in the specifications?

sample code http://scratchpad.io/descriptive-can-2849

Thank you,

I´m trying to pass the demo to a production server to make some test.
When i run the: 'npm run build' no errors show.
But if i run 'netstat -tnl | grep 8080' to see the services running on that port, nothing is running.
On dev its all OK