Question 1 of 30
In a speech recognition system designed for a multilingual application, the model needs to accurately transcribe spoken words into text across different languages. The system employs a hybrid approach that combines both acoustic models and language models. If the acoustic model achieves a word error rate (WER) of 10% for English and 15% for Spanish, while the language model improves the overall accuracy by reducing the WER by 50% for both languages, what will be the new WER for each language after applying the language model?
5% for English and 7.5% for Spanish
10% for English and 15% for Spanish
7.5% for English and 10% for Spanish
2.5% for English and 5% for Spanish

Preparing for Microsoft AI-100 Designing and Implementing an Azure AI Solution? Now land the interview.

73% of qualified candidates get rejected because of weak resumes. Build an ATS-optimized, recruiter-ready resume in under 5 minutes - free to start.

Build My Resume Free