Calling a phone number doesnt response with the answer.

[Sreekanth11reddy]

Ive updated the training index with custom data using json which is uploaded into Blob.
When i call the phone number and ask about a query, it keeps saying "Its taking me longer than expected. Thank you for your patience". I could see below logs in the report:

I know communication is not happening between app and the Search AI, but can anyone please help me on how to fix this problem.
Thanks.

[clemlesne]

Hello,

The data import script is not bundled with that demo, sorry.

You could use LlamaIndex, Semantic Memory, Scrape It Now, Azure Bring Your Own Data, … it’s up to you to choose what you prefer!

The data schema should meet the following model:

call-center-ai/app/models/training.py

Lines 8 to 16 in c4e8df2

	class TrainingModel(BaseModel, frozen=True):
	"""
	Represents a training document from AI Search.
	"""
	content: str
	id: UUID
	score: float
	title: str

Note that the index is automatically created for peace of mind:

call-center-ai/app/persistence/ai_search.py

Line 198 in c4e8df2

# Index configuration

[Sreekanth11reddy]

@clemlesne , Thanks a lot for quick response.

As iam pretty new to the suggested tools for data import, ive did it manually by following below steps:

1. Uploaded .json file to blob storage
   SCTraining (1).json
2. Azure Search Service-->Data Sources --> Add new Data Sources--> Maped it to .json uploaded to blob storage in step1
3. Azure Search Service-->Indexers-->Add new Index --> Selected already created "trainings" indexes and Data Sources created in step2
4. I now get some response when searched in Indexes
   

When i try to call phone number and ask the same question i used in Indexes Search(step4) i dont get response and bot keeps saying "Its taking me longer than expected. Thank you for your patience"

I also noticed a discrepancy in the "trainings" index — the fields include "vectors" but not "score." Is that expected?

Please do let me know if iam doing anything wrong.

[Sreekanth11reddy]

Updating the logs from container:2025-04-13T14:01:18.0107687Z stdout F 2025-04-13T14:01:18.010523Z [warning ] Hard timeout of 40s reached call.channel=voice call.id=c884235a-22fc-4334-9016-224d46331ac5 call.message=infram inframotion labs inframotion call.phone_number=+919481181498
2025-04-13T14:01:18.0108702Z stdout F 2025-04-13T14:01:18.010626Z [info ] Retrying chat, 1 remaining call.channel=voice call.id=c884235a-22fc-4334-9016-224d46331ac5 call.message=infram inframotion labs inframotion call.phone_number=+919481181498
2025-04-13T14:01:18.0200239Z stdout F 2025-04-13T14:01:18.019787Z [info ] Enhancing LLM chat with 0 trainings call.channel=voice call.id=c884235a-22fc-4334-9016-224d46331ac5 call.message=infram inframotion labs inframotion call.phone_number=+919481181498
2025-04-13T14:01:18.0524118Z stdout F 2025-04-13T14:01:18.052172Z [info ] Feature slow_llm_for_chat not found, using default: True call.channel=voice call.id=c884235a-22fc-4334-9016-224d46331ac5 call.message=infram inframotion labs inframotion call.phone_number=+919481181498
2025-04-13T14:01:18.0531494Z stdout F 2025-04-13T14:01:18.052958Z [info ] Using 13/13 messages (5530 tokens) as context call.channel=voice call.id=c884235a-22fc-4334-9016-224d46331ac5 call.message=infram inframotion labs inframotion call.phone_number=+919481181498
2025-04-13T14:01:33.0216137Z stdout F 2025-04-13T14:01:33.021357Z [warning ] Soft timeout of 15s reached call.channel=voice call.id=c884235a-22fc-4334-9016-224d46331ac5 call.message=infram inframotion labs inframotion call.phone_number=+919481181498
2025-04-13T14:01:33.0217918Z stdout F 2025-04-13T14:01:33.021601Z [info ] Playing TTS: It's taking me longer than expected to reply. Thank you for your patience… call.channel=voice call.id=c884235a-22fc-4334-9016-224d46331ac5 call.message=infram inframotion labs inframotion call.phone_number=+919481181498
2025-04-13T14:01:58.0432008Z stdout F 2025-04-13T14:01:58.042893Z [warning ] Hard timeout of 40s reached call.channel=voice call.id=c884235a-22fc-4334-9016-224d46331ac5 call.message=infram inframotion labs inframotion call.phone_number=+919481181498
2025-04-13T14:01:58.0432880Z stdout F 2025-04-13T14:01:58.043011Z [info ] Retrying chat, 0 remaining call.channel=voice call.id=c884235a-22fc-4334-9016-224d46331ac5 call.message=infram inframotion labs inframotion call.phone_number=+919481181498
2025-04-13T14:01:58.0500956Z stdout F 2025-04-13T14:01:58.049901Z [info ] Enhancing LLM chat with 0 trainings call.channel=voice call.id=c884235a-22fc-4334-9016-224d46331ac5 call.message=infram inframotion labs inframotion call.phone_number=+919481181498
2025-04-13T14:01:58.0511136Z stdout F 2025-04-13T14:01:58.050943Z [warning ] Tools disabled for this chat call.channel=voice call.id=c884235a-22fc-4334-9016-224d46331ac5 call.message=infram inframotion labs inframotion call.phone_number=+919481181498
2025-04-13T14:01:58.0518722Z stdout F 2025-04-13T14:01:58.051631Z [info ] Feature slow_llm_for_chat not found, using default: True call.channel=voice call.id=c884235a-22fc-4334-9016-224d46331ac5 call.message=infram inframotion labs inframotion call.phone_number=+919481181498
2025-04-13T14:01:58.0522786Z stdout F 2025-04-13T14:01:58.052142Z [info ] Using 13/13 messages (2285 tokens) as context call.channel=voice call.id=c884235a-22fc-4334-9016-224d46331ac5 call.message=infram inframotion labs inframotion call.phone_number=+919481181498
2025-04-13T14:02:13.0447406Z stdout F 2025-04-13T14:02:13.044462Z [warning ] Soft timeout of 15s reached call.channel=voice call.id=c884235a-22fc-4334-9016-224d46331ac5 call.message=infram inframotion labs inframotion call.phone_number=+919481181498
2025-04-13T14:02:13.0449250Z stdout F 2025-04-13T14:02:13.044708Z [info ] Playing TTS: Please hold on, I'm almost done. call.channel=voice call.id=c884235a-22fc-4334-9016-224d46331ac5 call.message=infram inframotion labs inframotion call.phone_number=+919481181498

[Sreekanth11reddy]

@clemlesne , just an update. Ive used LlamaIndex to import the data and i still see no response. Vector Quota is still 0

I still set no response when i manually search:

Any suggestion would really help me. Thanks

[Sreekanth11reddy]

@clemlesne ,Can you please give any suggestion. Thanks