MLS-C01 · Question #187
A company wants to use automatic speech recognition (ASR) to transcribe messages that are less than 60 seconds long from a voicemail-style application. The company requires the correct…
The correct answer is C. Create a custom vocabulary file containing each product name with phonetic. Amazon Transcribe Custom Vocabulary is the correct tool here. A custom vocabulary file allows you to list domain-specific terms - including unusual product names - along with optional phonetic pronunciations (using IPA or SoundsLike fields) and display forms. This directly…
Question
A company wants to use automatic speech recognition (ASR) to transcribe messages that are less than 60 seconds long from a voicemail-style application. The company requires the correct identification of 200 unique product names, some of which have unique spellings or pronunciations. The company has 4,000 words of Amazon SageMaker Ground Truth voicemail transcripts it can use to customize the chosen ASR model. The company needs to ensure that everyone can update their customizations multiple times each hour. Which approach will maximize transcription accuracy during the development phase?
Options
- AUse a voice-driven Amazon Lex bot to perform the ASR customization. Create customer
- BUse Amazon Transcribe to perform the ASR customization. Analyze the word confidence
- CCreate a custom vocabulary file containing each product name with phonetic
- DUse the audio transcripts to create a training dataset and build an Amazon Transcribe
How the community answered
(37 responses)- A8% (3)
- B14% (5)
- C76% (28)
- D3% (1)
Explanation
Amazon Transcribe Custom Vocabulary is the correct tool here. A custom vocabulary file allows you to list domain-specific terms - including unusual product names - along with optional phonetic pronunciations (using IPA or SoundsLike fields) and display forms. This directly solves the problem of unique spellings and pronunciations. Custom vocabularies can be created and updated via API, supporting multiple updates per hour as required. Option A (Amazon Lex) is a conversational AI service, not an ASR transcription service. Option B (analyzing word confidence scores) is a reactive diagnostic step, not a proactive accuracy improvement. Option D (training a custom Amazon Transcribe model) requires significantly more than 4,000 words of data and takes much longer to develop - it does not support rapid, frequent updates during a development phase.
Topics
Community Discussion
No community discussion yet for this question.