ExamTopics Logo
Mail Us[email protected]
Menu
Amazon Discussions
exam questions

Exam AWS Certified Machine Learning - Specialty All Questions

View all questions & answers for the AWS Certified Machine Learning - Specialty exam
Go to Exam

Exam AWS Certified Machine Learning - Specialty topic 1 question 161 discussion

Exam question from Amazon's AWS Certified Machine Learning - Specialty
Question #: 161
Topic #: 1
[All AWS Certified Machine Learning - Specialty Questions]

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?

  • A. Use a voice-driven Amazon Lex bot to perform the ASR customization. Create customer slots within the bot that specifically identify each of the required product names. Use the Amazon Lex synonym mechanism to provide additional variations of each product name as mis-transcriptions are identified in development.
  • B. Use Amazon Transcribe to perform the ASR customization. Analyze the word confidence scores in the transcript, and automatically create or update a custom vocabulary file with any word that has a confidence score below an acceptable threshold value. Use this updated custom vocabulary file in all future transcription tasks.
  • C. Create a custom vocabulary file containing each product name with phonetic pronunciations, and use it with Amazon Transcribe to perform the ASR customization. Analyze the transcripts and manually update the custom vocabulary file to include updated or additional entries for those names that are not being correctly identified.
  • D. Use the audio transcripts to create a training dataset and build an Amazon Transcribe custom language model. Analyze the transcripts and update the training dataset with a manually corrected version of transcripts where product names are not being transcribed correctly. Create an updated custom language model.
Show Suggested Answer Hide Answer
Suggested Answer: C 🗳️
by bluer1 at May 2, 2022, 4:10 p.m.
Disclaimers:
  • - ExamTopics website is not related to, affiliated with, endorsed or authorized by Amazon.
  • - Trademarks, certification & product names are used for reference only and belong to Amazon.

Comments

Chosen Answer:
This is a voting comment (?). It is better to Upvote an existing comment if you don't have anything to add.
Switch to a voting comment New
Submit Cancel
siju13
Highly Voted 2 years, 5 months ago
Selected Answer: C
Answer is C. https://aws.amazon.com/blogs/machine-learning/build-a-custom-vocabulary-to-enhance-speech-to-text-transcription-accuracy-with-amazon-transcribe/
upvoted 12 times
...
AjoseO
Highly Voted 1 year, 8 months ago
Selected Answer: D
Option D involves using the available audio transcripts to create a training dataset and building a custom language model with Amazon Transcribe. This approach provides a high degree of control over the transcription process and the ability to fine-tune the model to the specific vocabulary and pronunciation requirements of the company. Analyzing the transcripts and updating the training dataset with corrected versions is a crucial step in improving transcription accuracy. It enables the model to learn from mistakes and to incorporate the unique spelling and pronunciation of the 200 required product names.
upvoted 5 times
Siyuan_Zhu
1 year, 8 months ago
Thank you AjoseO for all these detailed explanations! They are very useful!
upvoted 2 times
ccpmad
1 year, 3 months ago
say thank you to chat gpt
upvoted 2 times
...
...
drcok87
1 year, 8 months ago
D is an ideal answer however, the question ask for "The company needs to ensure that everyone can update their customizations multiple times each hour". To retrain custom model each hour when we have changes, will be tedious and time consuming. I go with c, where we can ask everyone to just update the config file.
upvoted 8 times
...
...
Carpediem78
Most Recent 1 month ago
Selected Answer: D
The company requires the correct identification of 200 unique product names, some of which have unique spellings or pronunciations. -> Use the audio transcripts to create a training dataset and build an Amazon Transcribe custom language model. Analyze the transcripts and update the training dataset with a manually corrected version of transcripts where product names are not being transcribed correctly. Create an updated custom language model.
upvoted 1 times
ef12052
3 weeks, 6 days ago
why? i think it's C
upvoted 1 times
...
...
AIWave
8 months, 1 week ago
Selected Answer: C
-Creating a custom vocabulary file allows you to explicitly define the correct pronunciation of each product name. -Manually updating the custom vocabulary file based on these observations allows you to continuously improve the ASR system. - As new product names or variations emerge, you can easily add them to the custom vocabulary file without retraining the entire ASR model.
upvoted 2 times
...
Shenannigan
1 year, 2 months ago
Selected Answer: C
D was my initial choice however looking at the requirement "The company needs to ensure that everyone can update their customizations multiple times each hour." I changed my mind due to having to retrain the model with new vocabulary. C gives you the ability to update the vocabulary and have it take effect immediately
upvoted 2 times
...
kaike_reis
1 year, 2 months ago
Selected Answer: C
Answer is C D would required to build a model. It's well known the quantity of products, so it's not necessary.
upvoted 1 times
...
Mickey321
1 year, 3 months ago
Selected Answer: D
the best approach to maximize transcription accuracy during the development phase is to use the audio transcripts to create a training dataset and build an Amazon Transcribe custom language model. Analyze the transcripts and update the training dataset with a manually corrected version of transcripts where product names are not being transcribed correctly. Create an updated custom language model.
upvoted 1 times
...
DeepakPg
1 year, 10 months ago
Why not D though?
upvoted 1 times
...
Yongs
1 year, 10 months ago
I think C is correct.
upvoted 2 times
...
bluer1
2 years, 6 months ago
A? any thought?
upvoted 1 times
...
Community vote distribution
A (35%)
C (25%)
B (20%)
Other
Most Voted
A voting comment increases the vote count for the chosen answer by one.

Upvoting a comment with a selected answer will also increase the vote count towards that answer by one. So if you see a comment that you already agree with, you can upvote it instead of posting a new comment.

SaveCancel
exam
Someone Bought Contributor Access for:
SY0-701
London, 1 minute ago