Skip to content

Commit 02a467c

Browse files
eric-urbaneric-urban
committed
clu output json detail
1 parent 7b1436f commit 02a467c

File tree

3 files changed

+159
-6
lines changed

3 files changed

+159
-6
lines changed

‎articles/cognitive-services/Speech-Service/includes/quickstarts/intent-recognition-clu/cpp.md

Lines changed: 79 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -26,7 +26,7 @@ The Speech SDK is available as a [NuGet package](https://www.nuget.org/packages/
2626

2727
[!INCLUDE [Deploy CLU model](deploy-clu-model.md)]
2828

29-
You'll need the project name and deployment name in the next section.
29+
You'll use the project name and deployment name in the next section.
3030

3131
## Recognize intents from a microphone
3232

@@ -156,6 +156,84 @@ RECOGNIZED: Text=Go ahead and delete the e-mail.
156156
Language Understanding JSON: {"kind":"ConversationResult","result":{"query":"go ahead and delete the email","prediction":{"topIntent":"Delete","projectKind":"Conversation","intents":[{"category":"Delete","confidenceScore":0.95413816},{"category":"Cancel","confidenceScore":0.8985081},{"category":"SendEmail","confidenceScore":0.76236105},{"category":"Confirm","confidenceScore":0.7585944},{"category":"QueryLastText","confidenceScore":0.7434454},{"category":"CheckMessages","confidenceScore":0.70770377},{"category":"SearchMessages","confidenceScore":0.7021245},{"category":"ReadAloud","confidenceScore":0.6168782},{"category":"AddMore","confidenceScore":0.6116944},{"category":"Reply","confidenceScore":0.6059992},{"category":"ShowNext","confidenceScore":0.5359252},{"category":"AddFlag","confidenceScore":0.504695},{"category":"ShowPrevious","confidenceScore":0.49446106},{"category":"Forward","confidenceScore":0.41094807},{"category":"None","confidenceScore":0.34945187}],"entities":[]}}}.
157157
```
158158

159+
The intents are returned in the probability order of most likely to least likely. Here's a formatted version of the JSON output where the `topIntent` is `Delete` with a confidence score of 0.95413816 (95.41%):
160+
161+
```json
162+
{
163+
"kind": "ConversationResult",
164+
"result": {
165+
"query": "go ahead and delete the email",
166+
"prediction": {
167+
"topIntent": "Delete",
168+
"projectKind": "Conversation",
169+
"intents": [
170+
{
171+
"category": "Delete",
172+
"confidenceScore": 0.95413816
173+
},
174+
{
175+
"category": "Cancel",
176+
"confidenceScore": 0.8985081
177+
},
178+
{
179+
"category": "SendEmail",
180+
"confidenceScore": 0.76236105
181+
},
182+
{
183+
"category": "Confirm",
184+
"confidenceScore": 0.7585944
185+
},
186+
{
187+
"category": "QueryLastText",
188+
"confidenceScore": 0.7434454
189+
},
190+
{
191+
"category": "CheckMessages",
192+
"confidenceScore": 0.70770377
193+
},
194+
{
195+
"category": "SearchMessages",
196+
"confidenceScore": 0.7021245
197+
},
198+
{
199+
"category": "ReadAloud",
200+
"confidenceScore": 0.6168782
201+
},
202+
{
203+
"category": "AddMore",
204+
"confidenceScore": 0.6116944
205+
},
206+
{
207+
"category": "Reply",
208+
"confidenceScore": 0.6059992
209+
},
210+
{
211+
"category": "ShowNext",
212+
"confidenceScore": 0.5359252
213+
},
214+
{
215+
"category": "AddFlag",
216+
"confidenceScore": 0.504695
217+
},
218+
{
219+
"category": "ShowPrevious",
220+
"confidenceScore": 0.49446106
221+
},
222+
{
223+
"category": "Forward",
224+
"confidenceScore": 0.41094807
225+
},
226+
{
227+
"category": "None",
228+
"confidenceScore": 0.34945187
229+
}
230+
],
231+
"entities": []
232+
}
233+
}
234+
}
235+
```
236+
159237
## Remarks
160238
Now that you've completed the quickstart, here are some additional considerations:
161239

‎articles/cognitive-services/Speech-Service/includes/quickstarts/intent-recognition-clu/csharp.md

Lines changed: 79 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -26,7 +26,7 @@ The Speech SDK is available as a [NuGet package](https://www.nuget.org/packages/
2626

2727
[!INCLUDE [Deploy CLU model](deploy-clu-model.md)]
2828

29-
You'll need the project name and deployment name in the next section.
29+
You'll use the project name and deployment name in the next section.
3030

3131
## Recognize intents from a microphone
3232

@@ -141,6 +141,84 @@ RECOGNIZED: Text=Go ahead and delete the e-mail.
141141
Language Understanding JSON: {"kind":"ConversationResult","result":{"query":"go ahead and delete the email","prediction":{"topIntent":"Delete","projectKind":"Conversation","intents":[{"category":"Delete","confidenceScore":0.95413816},{"category":"Cancel","confidenceScore":0.8985081},{"category":"SendEmail","confidenceScore":0.76236105},{"category":"Confirm","confidenceScore":0.7585944},{"category":"QueryLastText","confidenceScore":0.7434454},{"category":"CheckMessages","confidenceScore":0.70770377},{"category":"SearchMessages","confidenceScore":0.7021245},{"category":"ReadAloud","confidenceScore":0.6168782},{"category":"AddMore","confidenceScore":0.6116944},{"category":"Reply","confidenceScore":0.6059992},{"category":"ShowNext","confidenceScore":0.5359252},{"category":"AddFlag","confidenceScore":0.504695},{"category":"ShowPrevious","confidenceScore":0.49446106},{"category":"Forward","confidenceScore":0.41094807},{"category":"None","confidenceScore":0.34945187}],"entities":[]}}}.
142142
```
143143

144+
The intents are returned in the probability order of most likely to least likely. Here's a formatted version of the JSON output where the `topIntent` is `Delete` with a confidence score of 0.95413816 (95.41%):
145+
146+
```json
147+
{
148+
"kind": "ConversationResult",
149+
"result": {
150+
"query": "go ahead and delete the email",
151+
"prediction": {
152+
"topIntent": "Delete",
153+
"projectKind": "Conversation",
154+
"intents": [
155+
{
156+
"category": "Delete",
157+
"confidenceScore": 0.95413816
158+
},
159+
{
160+
"category": "Cancel",
161+
"confidenceScore": 0.8985081
162+
},
163+
{
164+
"category": "SendEmail",
165+
"confidenceScore": 0.76236105
166+
},
167+
{
168+
"category": "Confirm",
169+
"confidenceScore": 0.7585944
170+
},
171+
{
172+
"category": "QueryLastText",
173+
"confidenceScore": 0.7434454
174+
},
175+
{
176+
"category": "CheckMessages",
177+
"confidenceScore": 0.70770377
178+
},
179+
{
180+
"category": "SearchMessages",
181+
"confidenceScore": 0.7021245
182+
},
183+
{
184+
"category": "ReadAloud",
185+
"confidenceScore": 0.6168782
186+
},
187+
{
188+
"category": "AddMore",
189+
"confidenceScore": 0.6116944
190+
},
191+
{
192+
"category": "Reply",
193+
"confidenceScore": 0.6059992
194+
},
195+
{
196+
"category": "ShowNext",
197+
"confidenceScore": 0.5359252
198+
},
199+
{
200+
"category": "AddFlag",
201+
"confidenceScore": 0.504695
202+
},
203+
{
204+
"category": "ShowPrevious",
205+
"confidenceScore": 0.49446106
206+
},
207+
{
208+
"category": "Forward",
209+
"confidenceScore": 0.41094807
210+
},
211+
{
212+
"category": "None",
213+
"confidenceScore": 0.34945187
214+
}
215+
],
216+
"entities": []
217+
}
218+
}
219+
}
220+
```
221+
144222
## Remarks
145223
Now that you've completed the quickstart, here are some additional considerations:
146224

‎articles/cognitive-services/Speech-Service/includes/quickstarts/intent-recognition-clu/deploy-clu-model.md

Lines changed: 1 addition & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -21,7 +21,7 @@ Once the upload is complete, you will land on **Schema definition** page. For th
2121

2222
### Train your model
2323

24-
Typically, after you create a project, you should [build a schema](../../how-to/build-schema.md) and [label utterances](../../how-to/tag-utterances.md). For this quickstart, we already imported a ready project with built schema and labeled utterances.
24+
Typically, after you create a project, you should [build a schema](../../../../language-service/conversational-language-understanding/how-to/build-schema.md) and [label utterances](../../../../language-service/conversational-language-understanding/how-to/tag-utterances.md). For this quickstart, we already imported a ready project with built schema and labeled utterances.
2525

2626
To train a model, you need to start a training job. The output of a successful training job is your trained model.
2727

@@ -33,7 +33,4 @@ Generally after training a model you would review its evaluation details. In thi
3333

3434
[!INCLUDE [Deploy model](../../../../language-service/conversational-language-understanding/includes/language-studio/deploy-model.md)]
3535

36-
### Test your model
37-
38-
[!INCLUDE [Test model](../../../../language-service/conversational-language-understanding/includes/language-studio/test-model.md)]
3936

0 commit comments

Comments
 (0)