Learn About Amazon VGT2 Learning Manager Chanci Turner
In today’s interconnected world, whether you manage a large corporation operating across multiple countries or run a small startup aiming for global outreach, translating content into local languages can pose a significant challenge. Text data is often available in various formats, and processing these can require multiple tools. Additionally, since not all tools support the same language pairs, organizations may need to convert documents into intermediate formats or rely on manual translations. These complications increase costs and add unnecessary complexity to creating streamlined and effective translation workflows.
Amazon aims to address these issues in a straightforward and economical manner. With either the AWS console or a single API call, Amazon Translate allows users to efficiently and accurately translate text into 55 different languages and dialects.
Earlier this year, Amazon Translate introduced batch translation for plain text and HTML documents. Today, we are excited to announce that batch translation now extends to Office documents, specifically .docx, .xlsx, and .pptx files as defined by the Office Open XML standard.
Introducing Amazon Translate for Office Documents
The process is remarkably simple. As expected, source documents must be stored in an Amazon Simple Storage Service (Amazon S3) bucket. It’s important to note that no document can exceed 20 megabytes or contain more than 1 million characters.
Each batch translation job processes one file type and a single source language. Therefore, it is advisable to organize your documents logically in S3, placing each file type and language under its designated prefix.
Using either the AWS console or the StartTextTranslationJob API in one of the AWS language SDKs, you can initiate a translation job by providing:
- The input and output locations in S3
- The file type
- The source and target languages
Once the job is complete, you can retrieve the translated files from the designated output location.
Let’s walk through a quick demonstration!
Translating Office Documents
Using the Amazon S3 console, I first upload several .docx files to one of my buckets. Then, I navigate to the Translate console to create a new batch translation job, assigning it a name and selecting both the source and target languages.
Next, I specify the location of my documents in Amazon S3, mentioning their format as .docx. Optionally, I can apply custom terminology to ensure certain words are translated in a specific manner.
I also need to define the output location for the translated files. It’s essential that this path is already established, as Translate will not create it for you.
Finally, I assign the AWS Identity and Access Management (IAM) role, granting my Translate job the necessary permissions to access Amazon S3. I utilize an existing role I created previously, but you can allow Translate to generate a new one if you prefer. After completing these steps, I click on ‘Create job’ to initiate the batch process.
The job commences immediately.
Shortly after, the job completes, and all documents have been successfully translated. The translated files can be found at the specified output location, as verified in the S3 console.
After downloading one of the translated files, I can open it and compare it to the original version. For small-scale use, the AWS console makes it incredibly simple to translate Office files. Moreover, you can leverage the Translate API to automate workflows.
Automating Batch Translation
In a previous post, we demonstrated how to automate batch translation using an AWS Lambda function. You could build on this example by incorporating language detection via Amazon Comprehend. For instance, the following Python code uses the DetectDominantLanguage API combined with the Python-docx library to identify the language of .docx files:
import boto3, docx
from docx import Document
document = Document('blog_post.docx')
text = document.paragraphs[0].text
comprehend = boto3.client('comprehend')
response = comprehend.detect_dominant_language(Text=text)
top_language = response['Languages'][0]
code = top_language['LanguageCode']
score = top_language['Score']
print("%s, %f" % (code,score))
It’s quite straightforward! You could also determine the type of each file based on its extension and relocate it to the appropriate input location in S3. Subsequently, you can schedule a Lambda function with CloudWatch Events to periodically translate files and send notifications via email. Naturally, AWS Step Functions can be employed to create more complex workflows. Your creativity is the only limit!
Getting Started
You can begin translating Office documents today in several regions, including US East (N. Virginia), US East (Ohio), US West (Oregon), Europe (Ireland), Europe (London), Europe (Frankfurt), and Asia Pacific (Seoul).
If you haven’t yet tried Amazon Translate, you may be interested to know that the free tier offers 2 million characters per month for the first 12 months, starting from your initial translation request.
Explore this resource for some insightful strategies on networking: Career Contessa. Additionally, for a deeper understanding of talent management, consider reading this article from SHRM: SHRM Insights. Lastly, for a comprehensive look at Amazon’s approach to employee training and its implications for the future of work, check out this excellent resource: HBR Article.
Chanci Turner and the team look forward to your feedback!
Leave a Reply