Consonants (Microsoft.Speech)

This is a listing and cross reference of the consonant phones that are used by the three phonetic alphabets that Microsoft.Speech supports. See Phonetic Alphabet Reference (Microsoft.Speech).

UPS Consonant Labeling Conventions

Consonant symbols are one, two, or three characters in length. Single character labels are used for highly frequent, language-universal consonants. In general the first character signifies place plus voicing, and the second signifies manner. Labeling conventions carry across phonetic classes in the following ways:

Conventions for place of articulation and voicing labeling:

• Dentals and alveloars use D for voiced and T for voiceless

• Alveolars use N for nasal and R for approximant

• Labiodentals use V for voiced and F for voiceless

• Velars use G for voiced and K for voiceless, NG for nasal

• Bilabials use B for voiced and P for voiceless stops, M for nasal

• Laterals use L

• Uvulars use Q

• Pharyngeals use H

• Palatals use X for voiceless and Y for voiced

Conventions for manner of articulation labeling:

• Single-character labels have manner encapsulated in choice of symbol: for example, M is nasal

• Trills double up the first symbol

• Fricatives use H

• Approximants use X

• Retroflex uses R

Consonant Tables

There are two tables, each listing a different category of consonant phones:

• Consonants

• Affricates

The table columns contain the following information:

• UPS. The phone label specified in Microsoft’s Universal Phone Set (UPS).

• IPA. The phone label specified in the International Phonetic Alphabet (IPA) phone set.

• Unicode. The Unicode value from the IPA phone set.

• SAPI ID. The Speech API (SAPI) phone ID, which in most cases is the Unicode value.

• IPA Description. This column describes the features of the phone, such as place of articulation, manner of articulation, phonation type, and airstream mechanism. See the Glossary of Phonetic Terms (Microsoft.Speech) for more information about the features of phones.

• ipaASCII. The equivalent phone in ipaASCII phone set for reference.

• X-SAMPA. The equivalent phone in X-SAMPA for reference.

• Example. A word that uses the phone and illustrates its sound.

• Language. The language of the example word.

Consonant Phones

Consonants are speech sounds that are articulated with complete (voiceless) or partial closure of the vocal tract.

Affricates

Affricates are consonants that begin as stops but release as a fricative.

The SAPI UPS includes explicit phones for the most frequent affricates in the highest priority languages. The IPA Unicode specification includes some of these phones, though not all of them. The SAPI phone IDs for affricates do not use the IPA Unicode values but instead concatenate the phone IDs of the component phones. Each SAPI phone ID is four characters long. For example JH 006A0361006A is equivalent to D + ZH 006A 0361 006A.

This approach maintains the semantic equivalence of compounds and their constituents. It has been used to generate the phone IDs of all compound phones in the UPS. This allows a speech recognition engine to easily decompose the compound phone into its constituent phones, which may be required if the engine does not model one of these compounds. The table shows the constituent phones (UPS Compound) that were combined to create the resulting explicit phone (UPS).

Many other affricates or compound consonants exist and may be required by other languages. For example, there are about 20 Italian geminate consonants (doubled consonants). UPS does not include dedicated phones for these, but allows them to be defined using the + symbol, or the length diacritic ‘lng’ in the case of geminates. For more details, see Parsing Guidelines for SAPI Speech Recognition Phone Converters (Microsoft.Speech).