Skip to content

Commit 643a3e7

Browse files
JamesJBarnettJamesJBarnett
authored
Merge pull request #211050 from eric-urban/eur/markdown-no-loc
no-loc phonemes
2 parents c1e24a6 + 9a0e478 commit 643a3e7

File tree

4 files changed

+54
-55
lines changed

4 files changed

+54
-55
lines changed

‎articles/cognitive-services/Speech-Service/customize-pronunciation.md

Lines changed: 3 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -13,15 +13,16 @@ ms.custom: template-concept, ignite-fall-2021
1313

1414
You can specify the phonetic pronunciation of words using the Universal Phone Set (UPS) in a [structured text data](how-to-custom-speech-test-and-train.md#structured-text-data-for-training) file. The UPS is a machine-readable phone set that is based on the International Phonetic Set Alphabet (IPA). The IPA is a standard used by linguists world-wide.
1515

16-
UPS pronunciations consist of a string of UPS phones, each separated by whitespace. UPS phone labels are all defined using ASCII character strings.
16+
UPS pronunciations consist of a string of UPS phonemes, each separated by whitespace. UPS phoneme labels are all defined using ASCII character strings.
1717

1818
For steps on implementing UPS, see [Structured text phonetic pronunciation](how-to-custom-speech-test-and-train.md#structured-text-data-for-training). Structured text phonetic pronunciation data is separate from [pronunciation data](how-to-custom-speech-test-and-train.md#pronunciation-data-for-training), and they cannot be used together. The first one is "sounds-like" or spoken-form data, and is input as a separate file, and trains the model what the spoken form sounds like
1919

2020
[Structured text phonetic pronunciation data](how-to-custom-speech-test-and-train.md#structured-text-data-for-training) is specified per syllable in a markdown file. Separately, [pronunciation data](how-to-custom-speech-test-and-train.md#pronunciation-data-for-training) it input on its own, and trains the model what the spoken form sounds like. You can either use a pronunciation data file on its own, or you can add pronunciation within a structured text data file. The Speech service doesn't support training a model with both of those datasets as input.
2121

2222
See the sections in this article for the Universal Phone Set for each locale.
2323

24-
## en-US
24+
## :::no-loc text="en-US":::
25+
2526
[!INCLUDE [en-US](./includes/phonetic-sets/speech-to-text/en-us.md)]
2627

2728
## Next steps

‎articles/cognitive-services/Speech-Service/includes/phonetic-sets/speech-to-text/en-us.md

Lines changed: 48 additions & 50 deletions
Original file line numberDiff line numberDiff line change
@@ -9,58 +9,56 @@ ms.author: eur
99

1010
#### Consonants
1111

12-
| UPS Phones | IPA | Example |
12+
| UPS Phonemes | IPA | Example |
1313
|------------|---------|----------|
14-
| B | b | big |
15-
| CH | t.ʃ / ʧ | chin |
16-
| D | d | dig |
17-
| DH | ð | then |
18-
| DX | ɾ | butter |
19-
| F | f | fork |
20-
| G | g | gut |
21-
| H | h | help |
22-
| JH | d.ʒ / ʤ | joy |
23-
| K | k | cut |
24-
| L | l | lid |
25-
| M | m | mat |
26-
| N | n | no |
27-
| NG | ŋ | sing |
28-
| P | p | put |
29-
| R | ɻ | red |
30-
| S | s | sit |
31-
| SH | ʃ | she |
32-
| T | t | talk |
33-
| TH | θ | thin |
34-
| V | v | vat |
35-
| W | w | with |
36-
| J | j | yard |
37-
| Z | z | zap |
38-
| ZH | ʒ | pleasure |
14+
| `B` | b | :::no-loc text="big"::: |
15+
| `CH` | t.ʃ / ʧ | :::no-loc text="chin"::: |
16+
| `D` | d | :::no-loc text="dig"::: |
17+
| `DH` | ð | :::no-loc text="then"::: |
18+
| `DX` | ɾ | :::no-loc text="butter"::: |
19+
| `F` | f | :::no-loc text="fork"::: |
20+
| `G` | g | :::no-loc text="gut"::: |
21+
| `H` | h | :::no-loc text="help"::: |
22+
| `JH` | d.ʒ / ʤ | :::no-loc text="joy"::: |
23+
| `K` | k | :::no-loc text="cut"::: |
24+
| `L` | l | :::no-loc text="lid"::: |
25+
| `M` | m | :::no-loc text="mat"::: |
26+
| `N` | n | :::no-loc text="no"::: |
27+
| `NG` | ŋ | :::no-loc text="sing"::: |
28+
| `P` | p | :::no-loc text="put"::: |
29+
| `R` | ɻ | :::no-loc text="red"::: |
30+
| `S` | s | :::no-loc text="sit"::: |
31+
| `SH` | ʃ | :::no-loc text="she"::: |
32+
| `T` | t | :::no-loc text="talk"::: |
33+
| `TH` | θ | :::no-loc text="thin"::: |
34+
| `V` | v | :::no-loc text="vat"::: |
35+
| `W` | w | :::no-loc text="with"::: |
36+
| `J` | j | :::no-loc text="yard"::: |
37+
| `Z` | z | :::no-loc text="zap"::: |
38+
| `ZH` | ʒ | :::no-loc text="pleasure"::: |
3939

4040
### Vowels
4141

42-
| UPS Phone | IPA | Example |
42+
| UPS Phonemes | IPA | Example |
4343
|-----------|-----|-----------|
44-
| AA | ɑ | father |
45-
| AE | æ | cat |
46-
| AH | ʌ | cut |
47-
| AO | ɔ | dog |
48-
| AOX | ɔ.ə | four |
49-
| AU | ɑ.ʊ | foul |
50-
| AX | ə | ago |
51-
| AX rho | ɚ | minor |
52-
| AI | ɑ.ɪ | bite |
53-
| EH | ɛ | pet |
54-
| EHX | ɛ.ə | stairs |
55-
| ER | ɜ | fur |
56-
| ER rho | ɝ | urban |
57-
| EI | e.ɪ | ate |
58-
| IH | ɪ | fill |
59-
| I | i | feel |
60-
| O | o | go |
61-
| OI | ɔ.ɪ | toy |
62-
| OWX | o.ə | boa |
63-
| Q | ɒ | hot |
64-
| UH | ʊ | book |
65-
| U | u | too, blue |
66-
| UWX | u.ə | lure |
44+
| `AA` | ɑ | :::no-loc text="father"::: |
45+
| `AE` | æ | :::no-loc text="cat"::: |
46+
| `AH` | ʌ | :::no-loc text="cut"::: |
47+
| `AO` | ɔ | :::no-loc text="dog"::: |
48+
| `AOX` | ɔ.ə | :::no-loc text="four"::: |
49+
| `AU` | ɑ.ʊ | :::no-loc text="foul"::: |
50+
| `AX` | ə | :::no-loc text="ago"::: |
51+
| `AI` | ɑ.ɪ | :::no-loc text="bite"::: |
52+
| `EH` | ɛ | :::no-loc text="pet"::: |
53+
| `EHX` | ɛ.ə | :::no-loc text="stairs"::: |
54+
| `ER` | ɜ | :::no-loc text="fur"::: |
55+
| `EI` | e.ɪ | :::no-loc text="ate"::: |
56+
| `IH` | ɪ | :::no-loc text="fill"::: |
57+
| `I` | i | :::no-loc text="feel"::: |
58+
| `O` | o | :::no-loc text="go"::: |
59+
| `OI` | ɔ.ɪ | :::no-loc text="toy"::: |
60+
| `OWX` | o.ə | :::no-loc text="boa"::: |
61+
| `Q` | ɒ | :::no-loc text="hot"::: |
62+
| `UH` | ʊ | :::no-loc text="book"::: |
63+
| `U` | u | :::no-loc text="too":::, :::no-loc text="blue"::: |
64+
| `UWX` | u.ə | :::no-loc text="lure"::: |

‎articles/cognitive-services/Speech-Service/includes/phonetic-sets/text-to-speech/en-us.md

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -2,8 +2,8 @@
22

33
|Example 1 (onset for consonant, word-initial for vowel)|Example 2 (intervocalic for consonant, word-medial nucleus for vowel)|Example 3 (coda for consonant, word-final for vowel)|Comments|
44
|--|--|--|--|
5-
| burger /b er **1** r - g ax r/ | falafel /f ax - l aa **1** - f ax l/ | guitar /g ih - t aa **1** r/ | The Speech service phone set puts stress after the vowel of the stressed syllable. |
6-
| inopportune /ih **2** - n aa - p ax r - t uw 1 n/ | dissimilarity /d ih - s ih **2**- m ax - l eh 1 - r ax - t iy/ | workforce /w er 1 r k - f ao **2** r s/ | The Speech service phone set puts stress after the vowel of the sub-stressed syllable. |
5+
| :::no-loc text="burger"::: /b er **1** r - g ax r/ | :::no-loc text="falafel"::: /f ax - l aa **1** - f ax l/ | :::no-loc text="guitar"::: /g ih - t aa **1** r/ | The Speech service phone set puts stress after the vowel of the stressed syllable. |
6+
| :::no-loc text="inopportune"::: /ih **2** - n aa - p ax r - t uw 1 n/ | :::no-loc text="dissimilarity"::: /d ih - s ih **2**- m ax - l eh 1 - r ax - t iy/ | :::no-loc text="workforce"::: /w er 1 r k - f ao **2** r s/ | The Speech service phone set puts stress after the vowel of the sub-stressed syllable. |
77

88
#### English vowels
99

‎articles/cognitive-services/Speech-Service/speech-ssml-phonetic-sets.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -60,7 +60,7 @@ See the sections in this article for the phonemes that are specific to each loca
6060
## en-GB/en-IE/en-AU
6161
[!INCLUDE [en-GB](./includes/phonetic-sets/text-to-speech/en-gb.md)]
6262

63-
## en-US/en-CA
63+
## :::no-loc text="en-US/en-CA":::
6464
[!INCLUDE [en-US](./includes/phonetic-sets/text-to-speech/en-us.md)]
6565

6666
## es-ES

0 commit comments

Comments
 (0)