Skip to content

Commit 072d500

Browse files
Merge pull request #263197 from sally-baolian/patch-188
Update how-to-pronunciation-assessment.md
2 parents bc22bef + 300e602 commit 072d500

File tree

1 file changed

+9
-4
lines changed

1 file changed

+9
-4
lines changed

articles/ai-services/speech-service/how-to-pronunciation-assessment.md

Lines changed: 9 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -22,6 +22,8 @@ zone_pivot_groups: programming-languages-ai-services
2222
In this article, you learn how to evaluate pronunciation with speech to text through the Speech SDK. To [get pronunciation assessment results](#get-pronunciation-assessment-results), you apply the `PronunciationAssessmentConfig` settings to a `SpeechRecognizer` object.
2323

2424
> [!NOTE]
25+
> For information about availability of pronunciation assessment, see [supported languages](language-support.md?tabs=pronunciation-assessment) and [available regions](regions.md#speech-service).
26+
>
2527
> As a baseline, usage of pronunciation assessment costs the same as speech to text for pay-as-you-go or commitment tier [pricing](https://azure.microsoft.com/pricing/details/cognitive-services/speech-services). If you [purchase a commitment tier](../commitment-tier.md) for speech to text, the spend for pronunciation assessment goes towards meeting the commitment.
2628
>
2729
> For pricing differences between scripted and unscripted assessment, see [the pricing note](./pronunciation-assessment-tool.md#pricing).
@@ -615,13 +617,12 @@ You can get pronunciation assessment scores for:
615617
- Syllable groups
616618
- Phonemes in [SAPI](/previous-versions/windows/desktop/ee431828(v=vs.85)#american-english-phoneme-table) or [IPA](https://en.wikipedia.org/wiki/IPA) format
617619

618-
> [!NOTE]
619-
> The syllable group, phoneme name, and spoken phoneme of pronunciation assessment are currently only available for the en-US locale. For information about availability of pronunciation assessment, see [supported languages](language-support.md?tabs=pronunciation-assessment) and [available regions](regions.md#speech-service).
620-
621620
## Syllable groups
622621

623622
Pronunciation assessment can provide syllable-level assessment results. Grouping in syllables is more legible and aligned with speaking habits, as a word is typically pronounced syllable by syllable rather than phoneme by phoneme.
624623

624+
Pronunciation assessment supports syllable groups in `en-US` with IPA and in both `en-US` and `en-GB` with SAPI.
625+
625626
The following table compares example phonemes with the corresponding syllables.
626627

627628
| Sample word | Phonemes | Syllables |
@@ -635,7 +636,9 @@ To request syllable-level results along with phonemes, set the granularity [conf
635636

636637
## Phoneme alphabet format
637638

638-
For the `en-US` locale, the phoneme name is provided together with the score, to help identify which phonemes were pronounced accurately or inaccurately. For other locales, you can only get the phoneme score.
639+
Pronunciation assessment supports phoneme name in `en-US` with IPA and in `en-US`, `en-GB` and `zh-CN` with SAPI.
640+
641+
For locales that support phoneme name, the phoneme name is provided together with the score, to help identify which phonemes were pronounced accurately or inaccurately. For other locales, you can only get the phoneme score.
639642

640643
The following table compares example SAPI phonemes with the corresponding IPA phonemes.
641644

@@ -713,6 +716,8 @@ pronunciationAssessmentConfig?.phonemeAlphabet = "IPA"
713716

714717
With spoken phonemes, you can get confidence scores indicating how likely the spoken phonemes matched the expected phonemes.
715718

719+
Pronunciation assessment supports spoken phonemes in `en-US` with IPA and in both `en-US` and `en-GB` with SAPI.
720+
716721
For example, to obtain the complete spoken sound for the word "Hello", you can concatenate the first spoken phoneme for each expected phoneme with the highest confidence score. In the following assessment result, when you speak the word "hello", the expected IPA phonemes are "h ɛ l oʊ". However, the actual spoken phonemes are "h ə l oʊ". You have five possible candidates for each expected phoneme in this example. The assessment result shows that the most likely spoken phoneme was `"ə"` instead of the expected phoneme `"ɛ"`. The expected phoneme `"ɛ"` only received a confidence score of 47. Other potential matches received confidence scores of 52, 17, and 2.
717722

718723
```json

0 commit comments

Comments
 (0)