Skip to content

Commit ac9a60f

Browse files
committed
azure ai foundry references
1 parent 8501db6 commit ac9a60f

File tree

7 files changed

+13
-15
lines changed

7 files changed

+13
-15
lines changed

articles/ai-services/translator/custom-translator/azure-ai-foundry/beginners-guide.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -76,7 +76,7 @@ Finding in-domain quality data is often a challenging task that varies based on
7676

7777
## What is a BLEU score?
7878

79-
BLEU (Bilingual Evaluation Understudy) is an algorithm for evaluating the precision or accuracy of text that is machine translated from one language to another. custom translation uses the BLEU metric as one way of conveying translation accuracy.
79+
BLEU (Bilingual Evaluation Understudy) is an algorithm for evaluating the precision or accuracy of text that is machine translated from one language to another. Custom translation uses the BLEU metric as one way of conveying translation accuracy.
8080

8181
A BLEU score is a number between zero and 100. A score of zero indicates a low quality translation where nothing in the translation matched the reference. A score of 100 indicates a perfect translation that is identical to the reference. It's not necessary to attain a score of 100 - a BLEU score between 40 and 60 indicates a high-quality translation.
8282

articles/ai-services/translator/custom-translator/azure-ai-foundry/concepts/bleu-score.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -24,7 +24,7 @@ The BLEU algorithm compares consecutive phrases of the automatic translation wit
2424

2525
The BLEU score's strength is that it correlates well with human judgment. BLEU averages out individual sentence judgment errors over a test corpus, rather than attempting to devise the exact human judgment for every sentence.
2626

27-
A more extensive discussion of BLEU scores *see* [Microsoft Translator Hub - Discussion of BLEU Score](https://youtu.be/-UqDljMymMg). BLEU results depend strongly on the breadth of your domain; consistency of test, training and tuning data; and how much data you have available for training. If your models are trained within a narrow domain, and your training data is consistent with your test data, you can expect a high BLEU score.
27+
For a more extensive discussion of BLEU scores, *see* [Microsoft Translator Hub - Discussion of BLEU Score](https://youtu.be/-UqDljMymMg). BLEU results depend strongly on the breadth of your domain; consistency of test, training and tuning data; and how much data you have available for training. If your models are trained within a narrow domain, and your training data is consistent with your test data, you can expect a high BLEU score.
2828

2929
>[!NOTE]
3030
>A comparison between BLEU scores is only justifiable when BLEU results are compared with the same Test set, the same language pair, and the same MT engine. A BLEU score from a different test set is bound to be different.

articles/ai-services/translator/custom-translator/azure-ai-foundry/concepts/data-filtering.md

Lines changed: 4 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,7 @@
11
---
22
title: Azure AI Foundry custom translation data filtering
33
titleSuffix: Azure AI services
4-
description: How custom transltation training documents for a custom system undergo a series of processing and filtering steps.
4+
description: How custom translation training documents for a custom system undergo a series of processing and filtering steps.
55
author: laujan
66
manager: nitinme
77
ms.service: azure-ai-translator
@@ -18,13 +18,13 @@ When you submit documents to be used for training, the documents undergo a serie
1818

1919
## Sentence alignment
2020

21-
If your document isn't in XLIFF, `TMX`, or ALIGN format, custom translation aligns the sentences of your source and target documents to each other, sentence by sentence. custom translation doesn't perform document alignment – it follows your naming of the documents to find the matching document of the other language. Within the document, custom translation tries to find the corresponding sentence in the other language. It uses document markup like embedded HTML tags to help with the alignment.
21+
If your document isn't in XLIFF, `TMX`, or ALIGN format, custom translation aligns the sentences of your source and target documents to each other, sentence by sentence. Custom translation doesn't perform document alignment – it follows your naming of the documents to find the matching document of the other language. Within the document, custom translation tries to find the corresponding sentence in the other language. It uses document markup like embedded HTML tags to help with the alignment.
2222

23-
If you see a large discrepancy between the number of sentences in the source and target documents, your documents can't be parallel. The document pairs with a large difference (>10%) of sentences on each side warrant a second look to make sure they're indeed parallel. custom translation shows a warning next to the document if the sentence count differs suspiciously.
23+
If you see a large discrepancy between the number of sentences in the source and target documents, your documents can't be parallel. The document pairs with a large difference (>10%) of sentences on each side warrant a second look to make sure they're indeed parallel. Custom translation shows a warning next to the document if the sentence count differs suspiciously.
2424

2525
## Deduplication
2626

27-
Custom translation removes the sentences that are present in test and tuning documents from training data. The removal happens dynamically inside of the training run, not in the data processing step. custom translation reports the sentence count to you in the project overview before such removal. Deduplication doesn't apply if you choose to upload your own test and tuning documents.
27+
Custom translation removes the sentences that are present in test and tuning documents from training data. The removal happens dynamically inside of the training run, not in the data processing step. Custom translation reports the sentence count to you in the project overview before such removal. Deduplication doesn't apply if you choose to upload your own test and tuning documents.
2828

2929
## Length filter
3030

articles/ai-services/translator/custom-translator/azure-ai-foundry/concepts/document-formats-naming-convention.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -52,4 +52,4 @@ Translation Memory files (`TMX`, `XLF`, `XLIFF`, `LCL`, `XLSX`) aren't required
5252
## Next steps
5353

5454
> [!div class="nextstepaction"]
55-
> [Learn about managing custom translation projects](workspace-and-project.md#what-is-a-custom-translation-project)
55+
> [Learn about managing custom translation projects](workspace-and-project.md#what-is-an-azure-ai-foundry-custom-translation-project)

articles/ai-services/translator/custom-translator/azure-ai-foundry/concepts/sentence-alignment.md

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,7 @@
11
---
22
title: Azure AI Foundry custom translation sentence pairing and alignment
33
titleSuffix: Azure AI services
4-
description: During the training execution, sentences present in parallel documents are paired or aligned. custom translation learns translations one sentence at a time, by reading a sentence and translating it. Then it aligns words and phrases in these two sentences to each other.
4+
description: During the training execution, sentences present in parallel documents are paired or aligned. Custom translation learns translations one sentence at a time, by reading a sentence and translating it. Then it aligns words and phrases in these two sentences to each other.
55
author: laujan
66
manager: nitinme
77
ms.service: azure-ai-translator
@@ -14,7 +14,7 @@ ms.custom: cogserv-non-critical-translator
1414

1515
# Azure AI Foundry custom translation sentence pairing and alignment
1616

17-
After documents are uploaded, sentences present in parallel documents are paired or aligned. custom translation reports the number of sentences it was able to pair as the Aligned Sentences in each of the data sets.
17+
After documents are uploaded, sentences present in parallel documents are paired or aligned. Custom translation reports the number of sentences it was able to pair as the Aligned Sentences in each of the data sets.
1818

1919
## Pairing and alignment process
2020

articles/ai-services/translator/custom-translator/azure-ai-foundry/concepts/workspace-and-project.md

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,7 @@
11
---
22
title: What is an Azure AI Foundry custom translation project?
33
titleSuffix: Azure AI services
4-
description: In This article, learn the differences about creating projects as well as project categories and labels for the custom translation service.
4+
description: In This article, learn the differences about creating projects, project categories, and labels for the custom translation service.
55
#services: cognitive-services
66
author: laujan
77
manager: nitinme
@@ -28,7 +28,7 @@ included in both language pairs. Each project has a Category ID associated with
2828

2929
The category identifies the domain – the area of terminology and style you want to use – for your language pair. Choose the category most relevant to your documents.
3030

31-
In the same project, you can create language pairs for the same language pair in different categories. custom translation prevents creation of a duplicate language pair with the same language pair and category. Applying a label to your language pair allows you to avoid this restriction. Don't use labels unless you're building translation systems for multiple clients, as adding a unique label to your language pair is reflected in your language pairs Category ID.
31+
In the same project, you can create language pairs for the same language pair in different categories. Custom translation prevents creation of a duplicate language pair with the same language pair and category. Applying a label to your language pair allows you to avoid this restriction. Don't use labels unless you're building translation systems for multiple clients, as adding a unique label to your language pair is reflected in your language pairs Category ID.
3232

3333
## Language pair labels
3434

articles/ai-services/translator/custom-translator/azure-ai-foundry/release-notes.md

Lines changed: 2 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -1,17 +1,15 @@
11
---
22
title: "Release notes - custom translation"
33
titleSuffix: Azure AI services
4-
description: custom translation releases, improvements, bug fixes, and known issues.
4+
description: Custom translation releases, improvements, bug fixes, and known issues.
55
author: laujan
66
manager: nitinme
77
ms.service: azure-ai-translator
8-
ms.custom:
9-
- ignite-2023
108
ms.date: 05/19/2025
119
ms.author: lajanuar
1210
ms.topic: release-notes
1311
---
14-
# custom translation release notes
12+
# Custom translation release notes
1513

1614
This page presents the latest feature, improvement, bug fix, and known issue release notes for custom translation service.
1715

0 commit comments

Comments
 (0)