By using AWS re:Post, you agree to the AWS re:Post Terms of Use
AWS re:Postre:Post

AWS transcribe issue

0

Hello, I'm using AWS transcribe to extract text from records I have a record in Arabic language but it also contains some English words and Names. When I create a Transcribe Job it skips the English words/Names or changes it to the wrong Arabic words. Can I have 2 languages in one transcribing job? also I added custom vocabulary for all records its not improving the performance

I found a low performance in Arabic records text extraction, especially in Names, and English words within the Arabic record. So How can I improve my performance?

Regards,

Topics
AnalyticsComputeMachine Learning & AI
Tags
AnalyticsHigh Performance ComputeAmazon TranscribeMachine Learning & AIAmazon Rekognition
Language
English
rePost-User-1281893
asked 2 years ago442 views
1 Answer
0

Hi,

Yes you can have 2 languages in one transcript job by using the following: https://docs.aws.amazon.com/AWSJavaScriptSDK/v3/latest/clients/client-transcribe/interfaces/starttranscriptionjobcommandinput.html#languageoptions

You can specify the language options (Arab and English) so they should be analyzed and identified.

For performance, not having any benchmarks, I d recommend to watch this reinvent video and verify whether the suggestions may work for you: https://youtu.be/QV31gi64M1w

Hope it helps and if does, I d appreciate answer to be accepted so that community can benefit for clarity, thanks ;)

profile picture
EXPERT
Antonio_Lagrotteria
answered 2 years ago
  • rePost-User-1281893
    2 years ago

    I'm trying to add my results on PCA so I noticed that custom vocabulary is empty for all records added in PCA how can I use cloud formation stack to add the custom voc?

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Guidelines for Answering Questions

Relevant content