azure-samples / cognitive-services-language-understanding Goto Github PK
View Code? Open in Web Editor NEWSamples for the Language Understanding Intelligent Service (LUIS)
License: MIT License
Samples for the Language Understanding Intelligent Service (LUIS)
License: MIT License
Hi
I have a requirement to do integration between the batch transcription and LUIS wherein I will pass the full transcriptions as such to LUIS and get the intent of the audio.
As far as I know we can pass the data for intent analysis to LUIS as a query which accepts only 500 characters.
So here comes the question is it possible to pass the full transcription from speech to text batch transcription API to LUIS for intent analysis or we have to feed the data in chunks to LUIS ?
If we feed the data in chunks how we will get the overall intent of the audio?
Please suggest on the best possible way to achieve this.
Several of the links in the live version of the landing page seem to be broken.
To reproduce:
Most of the links in those tables currently lead to 404 errors.
x
)- [x] bug report -> please search issues before submitting
- [ ] feature request
- [ ] documentation issue or request
- [ ] regression (a behavior that used to work and stopped in a new release)
Expected both to give personName Ben and Jhon
LUISGen {Your export json file} -cs {your C# file name}
Unhandled Exception: System.ArgumentException: Can not convert Object to String. at Newtonsoft.Json.Linq.JToken.op_Explicit(JToken value) at System.Dynamic.UpdateDelegates.UpdateAndExecute1[T0,TRet](CallSite site, T0 arg0) at CallSite.Target(Closure , CallSite , Object ) at LUISGen.Utils.EntityApply(JObject entity, Action
1 action) in D:\a\1\s\packages\LUISGen\src\Utils.cs:line 94
at System.Dynamic.UpdateDelegates.UpdateAndExecuteVoid5[T0,T1,T2,T3,T4](CallSite site, T0 arg0, T1 arg1, T2 arg2, T3 arg3, T4 arg4)
at LUISGen.CSharp.Entities(Object app, Writer w) in D:\a\1\s\packages\LUISGen\src\CSharp.cs:line 161
at System.Dynamic.UpdateDelegates.UpdateAndExecuteVoid3[T0,T1,T2](CallSite site, T0 arg0, T1 arg1, T2 arg2)
at LUISGen.CSharp.Generate(String description, Object app, String className, String space, String outPath) in D:\a\1\s\packages\LUISGen\src\CSharp.cs:line 289
at System.Dynamic.UpdateDelegates.UpdateAndExecuteVoid6[T0,T1,T2,T3,T4,T5](CallSite site, T0 arg0, T1 arg1, T2 arg2, T3 arg3, T4 arg4, T5 arg5)
at LUISGen.Program.Main(String[] args) in D:\a\1\s\packages\LUISGen\src\Program.cs:line 182`
The command must create a C# LUIS Recognizer class.
Microsoft.Bot.Builder.AI.Luis 4.11.1
Microsoft.Bot.Builder 4.11.1
Please provide us with the following information:
We are extracting datetimev2 entities via LUIS api using prebuilt model. In the json response, we are not able to deserialiize it due to additional values being returned. For eg: "10D" is added without the correct format of the json response body.
x
)- [x] bug report -> please search issues before submitting
- [ ] feature request
- [ ] documentation issue or request
- [ ] regression (a behavior that used to work and stopped in a new release)
Submitted line - to extract datetimev2 entities
Mannauðslausnir - RMKT,$63.87,From Feb 28, 2021, 12:00 AM to Mar 21, 2021, 2:15 PM,2021 - Mannauðslausnir - RMKT - RMKT - 10D,6,858 Impressions,$63.87,2020 - AutoBoost - Post Engagement,$155.13,From Feb 28, 2021, 12:00 AM to Mar 21, 2021, 2:15 PM,2020 - AutoBoost - Post Engagement - IS,4,095 Impressions,$12.76,2020 - AutoBoost - Post Engagement - IS,8,270 Impressions,$24.13,2020 - AutoBoost - Post Engagement - IS,8,008 Impressions,$21.98,2020 - AutoBoost - Post Engagement - IS,6,277 Impressions
The json response for datetimev2 we got
[{ 'type': 'datetimerange', 'values': [{ 'timex': '(2021-02-28T00:00,2021-03-21T14:15,PT518H)', 'resolution': [{ 'start': '2021-02-28 00:00:00', 'end': '2021-03-21 14:15:00' }] }] }, '10D', { 'type': 'datetimerange', 'values': [{ 'timex': '(2021-02-28T00:00,2021-03-21T14:15,PT518H)', 'resolution': [{ 'start': '2021-02-28 00:00:00', 'end': '2021-03-21 14:15:00' }] }] } ]
JSON response with correct format.
Windows 7, 8 or 10. Linux (which distribution). macOS (Yosemite? El Capitan? Sierra?)
Microsoft.Azure.CognitiveServices.Language.LUIS.Runtime V3.0.0 nuget
Thanks! We'll be in touch soon.
Are there any examples on how to turn on utterance normalisation in C# like on https://docs.microsoft.com/en-us/azure/cognitive-services/luis/luis-concept-utterance .
I am trying to do it in an Micorosoft chatbot using Azure bot services sdk v4.
Please provide us with the following information:
x
)- [ ] bug report -> please search issues before submitting
- [ ] feature request
- [ ] documentation issue or request
- [ ] regression (a behavior that used to work and stopped in a new release)
Uncaught (in promise) 400: BadArgument: Cannot find an unlabeled query with the given text.
at Object.handle (https://www.luis.ai/app.314c72959f3f3643f950.js:1:28902)
at Object.handle (https://www.luis.ai/app.314c72959f3f3643f950.js:1:28679)
at https://www.luis.ai/app.314c72959f3f3643f950.js:1:30640
at Generator.next ()
at https://www.luis.ai/vendors.314c72959f3f3643f950.js:31:1430
at new Promise ()
at l (https://www.luis.ai/vendors.314c72959f3f3643f950.js:31:1204)
at https://www.luis.ai/app.314c72959f3f3643f950.js:1:30581
Delete or accept the endpoint utterance at Review
Windows 10.
Thanks! We'll be in touch soon.
x
)- [X] bug report -> please search issues before submitting
- [ ] feature request
- [ ] documentation issue or request
- [ ] regression (a behavior that used to work and stopped in a new release)
https://westus.dev.cognitive.microsoft.com/docs/services/5890b47c39e2bb17b84a55ff/operations/5be313cec181ae720aa2b26c/console
Entering the right credentials to retrieve info of our registered applications
{
"error": {
"code": "BadArgument",
"message": "The source completed without providing data to receive."
}
}
Windows 10
I need help of LUIS API Team to solve this problem, I'm trying to use Bot Composer to publish an update with LUIS Service, but it fails, the reason: API of LUIS v2 and v3.0 preview.
Please provide us with the following information:
x
)- [x ] bug report -> please search issues before submitting
- [ ] feature request
- [ ] documentation issue or request
- [ ] regression (a behavior that used to work and stopped in a new release)
Use the sample output run the endpoint
Here is the sample curl
curl -X POST "https://westus2.api.cognitive.microsoft.com/language/:analyze-conversations?projectName=OptumTestIVR&deploymentName=production&api-version=2021-11-01-preview" -H "Ocp-Apim-Subscription-Key: xxxxxxxxxx" -H "Apim-Request-Id: xxxxxxxxxx" -H "Content-Type: application/json" -d "{\"verbose\":true,\"query\":\"YOUR_QUERY_HERE\",\"language\":\"QUERY_LANGUAGE_HERE\"}"
Technically I don't know what language is supposed to be or Query_language. I don't know if that is causing the error.
the issue seems to be the language param. I don't know what else to put besides english or eng
{
"error": {
"code": "BadRequest",
"message": "One or more errors occurred. Corresponding activity ID: ab2dc68a-d5c0-49f9-b621-c97a1f1ccb21.",
"innerError": {
"message": "Failed to call the Conversation project OptumTestIVR and deployment production.."
}
}
}
for it to work
Windows 7, 8 or 10. Linux (which distribution). macOS (Yosemite? El Capitan? Sierra?)
Thanks! We'll be in touch soon.
Please provide us with the following information:
x
)- [ ] bug report -> please search issues before submitting
- [x ] feature request
- [ ] documentation issue or request
- [ ] regression (a behavior that used to work and stopped in a new release)
Run endpoint
The output of the transaction as an example does not give the full entity information:
The different entities are within the category of AffirmENT but I can't resolve to them directly per the LIST. This is very different than LUIS.
A list key property would be ideal.
{
"query": "no",
"prediction": {
"topIntent": "Affirmation_Prompt",
"projectKind": "conversation",
"intents": [
{
"category": "Affirmation_Prompt",
"confidenceScore": 1
}
],
"entities": [
{
"category": "AffirmENT",
"text": "no",
"offset": 0,
"length": 2,
"confidenceScore": 1
}
]
}
}
Windows 7, 8 or 10. Linux (which distribution). macOS (Yosemite? El Capitan? Sierra?)
Thanks! We'll be in touch soon.
Please provide us with the following information:
x
)- [x] bug report -> please search issues before submitting
- [ ] feature request
- [ ] documentation issue or request
- [ ] regression (a behavior that used to work and stopped in a new release)
Create a LUIS application on luis.ai with at least 100 Intents and 100 Features (non-Global Phrase Lists) on luis.ai website
Create a new Phrase List (non-Global)
Notice that the Phrase List appears on the list correctly
Refresh the page
Notice that the new Phrase List is not appearing anymore
No logging messages, but 429 erros on the console all the time.
Since the Phrase List limit for Luis is 500, I should be able to create them normally. It's good to point out that the JSON file shows the Phrase Lists applied correctly to the intent, but it doesn't show on the screen. That blocks me from being able to work on the Luis app properly using the web version (luis.ai).
Windows, MacOs
10 Pro, Mojave 10.14.6
Luis apps with a smaller number of intents and phrase lists will work normally. With larger apps, it seems it bugs somehow. Just by training the app will give me 429 errors on the console, which seems weird.
Thanks! We'll be in touch soon.
Is your feature request related to a problem? Please describe.
When using the container version, you need to access web service. SDK Abstraction don't work as sdk URLs differ in webservice endpoints...
Azure-Samples/cognitive-services-speech-sdk#424 (comment)
The problem here is that prediction results are not the same...
Exemple in code :
Task pred;
if (_config.UseLocalServices)
{
pred = GetPredictionFromEndpointAsync(_config.LuisAPIKey, _config.LuisContainerUrl, _config.LuisAppId, _config.LuisSlotName, question.ToLowerInvariant());
}
else
{
IList<DynamicList> dynamicLists = _luisFeedDynamicListService.LuisEntries.Select(e => new DynamicList(e.Key, RequestListExtensions.FromLuisEntries(e.Value))).ToList();
var predictionRequest = new PredictionRequest(question.ToLowerInvariant(), null, null, dynamicLists);
pred = _luisClient.Prediction.GetSlotPredictionAsync(
Guid.Parse(_config.LuisAppId),
slotName: _config.LuisSlotName,
predictionRequest,
verbose: true,
showAllIntents: true,
log: true);
}
Describe the solution you'd like
The proper solution would be to change only the enpoint like this:
Version Azure
_luisClient = new LUISRuntimeClient(new ApiKeyServiceClientCredentials(_config.LuisAPIKey))
{
Endpoint = _config.LuisHostRegion
};
Version container
_luisClient = new LUISRuntimeClient(new MockCredentials())
{
Endpoint = "http://localhost:5010"
};
How can you have exactly the same result from webservice ?
An article here gave some caution to using the Bing spellchecker api for washing all of your user utterances into your bot channel.
Point being the entity could become unravled when the spellchecker is turned on? So if you have an obscure word style entity starblazingIT this might be reconstructed to star blazing IT. Thus harming the entity.
With that said the article was written in 2016 so perhaps this concern has been addressed but I can't find any documentation on it.
The documentation I did find on it gave an outlet for ignoring spelling mistakes and simply label utterances and the LUIS would learn proper spelling as well as typos... That is a loaded statement. What does that mean exactly and is luis really learning proper spelling or just adding the bad spellings in there as an alternative answer?
Because if that is the case then one would have to watch misspelled utterances and add in variations all of the time. Or would Luis overtime training learn that a particular word can have variations and be misspelled and to add that in with the correct intent model of utterances?
Here is a link to the article and quote below:
Ignore spelling mistakes
If you don't want to use the Bing Spell Check API v7 service, you can label utterances that have spelling mistakes so that LUIS can learn proper spelling as well as typos. This option requires more labeling effort than using a spell checker.
Overall what is the best practice of word misspelling correction for your luis chatbot application?
- [ ] bug report -> please search issues before submitting
- [x] feature request
- [ ] documentation issue or request
- [ ] regression (a behavior that used to work and stopped in a new release)
Would it be possible to add a LUIS Authoring Environment in region within Canada.
The following documentation page shows no LUIS authoring environment hosted within Canada.
The westus
is recommended for Canada, however that is not within Canada.
https://docs.microsoft.com/en-us/azure/cognitive-services/luis/luis-reference-regions#feedback
Not applicable
We should be able to select a LUIS authoring environment hosted within Canada.
Not applicable
This is a major blocker for our product.
Regulatory requirements dictate that must host our product entirely within Canada.
During the pilot / prototype stage we're fine for testing within westus
but this is a blocker for production launch.
Please advise.
Update sample of bot-integration-samples/hotel-finder to use Bot Framework V4. This repo is linked to from a primary documentation source so not appropriate to still be exclusively using an old version, especially given how much things have changed.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.