Amazon Transcribe table format specifying IPA

Question

I am processing a custom vocabulary table using the console at: https://console.aws.amazon.com/transcribe/

I've chosen a line from the example given
```
Phrase[TAB]SoundsLike[TAB]IPA[TAB]DisplayAs
C.L.I.[TAB][TAB]sɪ ɛl aɪ[TAB]CLI`
```
TAB is replaced with actual tabs and line endings ending with LF

Once processed for an en-GB or en-US custom vocabulary, it fails with the error:
`Validation error: File contains invalid characters or format in the IPA column. Error at line 2.`

If I modify the file to include extra spaces it succeeds, as in
```
Phrase[TAB]SoundsLike[TAB]IPA[TAB]DisplayAs
C.L.I.[TAB][TAB]s ɪ ɛ l aɪ[TAB]CLI
```
This error indicates that the examples given are incorrect or outdated. Is there another way to specify the syllables of a word by grouping characters together as in the examples?

AWS-User-5747670 · Accepted Answer

This was an issue with the webpage examples. The examples have now been updated and now show spaces between all characters. So the issue I was initially seeing was intended behaviour, and spaces are required.

SeattleGreg · Answer

The example download file that is provided also has examples without spaces for IPA:

```
Phrase	IPA	SoundsLike	DisplayAs
Los-Angeles			Los Angeles
F.B.I.	ɛ f b i aɪ		FBI
Etienne		eh-tee-en	
```

Amazon Transcribe table format specifying IPA

添加回答

相关内容