AI-102 Question 245
Select 2You are designing a solution to convert real-time speech from a multilingual conference into text using Azure AI Speech service. The solution should support multiple languages dynamically and include speaker identification to distinguish between different speakers. Which two features should you enable in your configuration?
- A
Enable language auto-detection in the Speech-to-Text service.
- B
Use the Custom Neural Voice feature for speaker identification.
- C
Enable speaker diarization in the Speech-to-Text service.
- D
Configure a custom language model for each language used in the conference.
- E
Enable profanity filtering in the Speech-to-Text service.
Show answer and explanation
Correct answers: A, C
Explanation
To meet the requirement for dynamic multilingual support, enabling language auto-detection ensures that the Speech-to-Text service can process multiple languages without predefining them. To identify different speakers in the transcription, speaker diarization must be enabled. Together, these features address the scenario's requirements effectively.
- A. Correct.
Enabling language auto-detection allows the Speech-to-Text service to dynamically recognize the spoken language, which is essential for a multilingual conference.
- B. Incorrect.
The Custom Neural Voice feature is used to create synthetic voices and is unrelated to speaker identification in this context.
- C. Correct.
Speaker diarization is necessary to distinguish between different speakers in the transcription, meeting the requirement for speaker identification.
- D. Incorrect.
Configuring a custom language model is useful for improving accuracy for a specific language or domain but is not dynamic and does not meet the requirement for supporting multiple languages dynamically.
- E. Incorrect.
Profanity filtering ensures compliance with content standards but is not relevant to the requirement for multilingual support or speaker identification.