Odia

Letters

Combining marks

Numbers

Punctuation

Symbols

Other

Vowel sounds

Consonant sounds

	labial	dental	alveolar	post- alveolar	retroflex	palatal	velar	glottal
stops	p b	t d			ʈ ɖ		k ɡ
aspirated	pʰ bʰ	tʰ dʰ			ʈʰ ɖʰ		kʰ ɡʰ
affricates				t͡ʃ d͡ʒ
aspirated				t͡ʃʰ d͡ʒʰ
fricatives	v		s					ɦ
nasals	m		n		ɳ	ɲ	ŋ
approximants	w		l		ɭ	j
trills/flaps			ɾ		ɽ
aspirated					ɽʰ

Vowel summary table

The following table summarises the main vowel to character assigments.

ⓘ represents the inherent vowel. Diacritics are added to the vowels to indicate nasalisation (not shown here).

For additional details see vowel_mappings.

Inherent vowel

କ kɔ U+0B15 ORIYA LETTER KA

ɔ following a consonant is not written, but is seen as an inherent part of the consonant letter, so kɔ is written by simply using the consonant letter. This vowel sound is transcribed as a.d

Odia uses  0B4D, called halant, (the Odia equivalent of the Sanskrit virama) to indicate that the inherent vowel is not pronounced after a consonant, eg. the following explicitly represents just the sound k.କ୍ k͓

Word-final consonants without a following inherent vowel use the halant, If there is no halant, the vowel is pronounced, eg. compare ଫୁଲ ଇ-ମେଲ୍

Consonant clusters also use this character, but if the cluster forms a conjunct then the virama is not rendered visibly (see clusters).

Vowels after consonants

Post-consonant vowels are written using 9 combining marks (vowel signs). There is 1 pre-base form, and 3 circumgraphs.

In principle, there are no multipart vowels, however the 2 circumgraphs are decomposed into 2 parts each.

Vowels have short lengths only, although there are vestigial orthographic letters for long sounds that now represent alternatives for the short sounds.

Vowels may be nasalised, using the candrabindu diacritic.

Pre-base vowel sign

କେ ke U+0B15 ORIYA LETTER KA + U+0B47: ORIYA VOWEL SIGN E

େ

The sound e is written using 0B47, which appears to the left of the base consonant letter or cluster.

This is a combining mark that is always stored after the base consonant. The rendering process places the glyph before the base consonant.

When an orthographic syllable begins with a consonant cluster that is rendered as a conjunct, the vowel sign is rendered before the start of the cluster, eg. fig_prebase shows 3 sets of consonant clusters, each followed by e when spoken, but the vowel sign appears to the left of each cluster.

Circumgraphs

କୋ ke U+0B15 ORIYA LETTER KA + U+0B4B: ORIYA VOWEL SIGN O

ୋ␣ ␣ୈ␣ୌ

Three vowels are produced by a single combining character with visually separate parts, that appear on different sides of the consonant onset.

show composition

ସ୍ତ୍ରୈଣ

All 3 of these circumgraphs can be written as a single character, or as two. See vs_encoding.

Nasalisation

ଁ

Vowels may be nasalised using ଁ or ଂ.

show composition

ମୁଁହ

Where 2 vowels appear together, the nasalisation sign is rendered above the second, eg. ଜ୍ୱାଇଁ

Standalone vowels

Odia represents standalone vowels using a set of independent vowel letters. The set includes a character to represent the inherent vowel sound, ɔ.

Vowel sounds to characters

Consonant summary table

The following table summarises the main consonant to character assignments.

For additional details see vowel_mappings.

Basic consonants

Whereas the table just above takes you from sounds to letters, the following simply lists the basic consonant letters (however, since the orthography is highly phonetic there is little difference in ordering).

The velar and palatal nasals only occur in homorganic clusters.d,406

କ୍ଷ is regarded as a letter of the alphabet.

wa and va

The letters 0B71 and 0B35 were added to Unicode version 4.

The subjoined forms of 0B71 and 0B2C may look the same. For a discussion of the possible historical relationship between these characters see Everson/Stonee02.

Observation: The Library of Congress transcription page says that when ବ [U+0B2C ORIYA LETTER BA] occurs as the second consonant of a consonant cluster (except when geminated), it is transliterated vloc. It appears, however, that it also keeps the b sound after the letters m and r.

0B35 is described by Wiktionary as "used sporadically for the phonetic Va/Wa as an alternative for the officially recognised letter ୱ, but has not gained widespread acceptance".

Repertoire extension using nukta

The sounds ɽ and ɽʰ are written by combining 0B3C with an existing consonant.

The nukta should always be typed and stored immediately after the consonant it modifies, and before any combining vowels or diacritics.

Unicode also has precomposed forms of these letters, but they decompose under Unicode Normalisation Form C (NFC). Therefore, the Unicode Standard recommends the use of the decomposed sequence.

The nukta may also be used to produce other non-native sounds. Wiktionary describes the following:

• 0B15 0B3C for q.
• 0B16 0B3C for x.
• 0B1D 0B3C for ʒ.
• 0B37 0B3C for ɻ.

Onsets

Clusters of consonant letters at the beginning of an orthographic syllable occur in Odia, and they are handled as described in the section clusters.

Special behaviours include handling of RA at the beginning of an orthographic syllable (see rconjuncts).

Finals

A syllable-final nasal sound can be written using 0B02, eg. ଜଂତୁ ଜଂଗଲ ଏବଂ

It is optional whether the nasal sound is written using anusvara or by using a conjunct. fig_anusvara shows two ways of writing ଅଂକ.u,488

A word-final h consonant can also be written using 0B03.n,54 (In the middle of a word, it creates a geminated consonant.)

Observation: According to Wikipedia, that sound is a h, but according to Nakanishi it is a glottal stop.

Consonant clusters

The absence of a vowel sound between two or more consonants is visually indicated in one of the following ways.

1. Create a conjunct. There are a number of possibilities here:
   1. Stacking : Reduce a non-initial consonant in size and shape and position it below the first.
   2. Conjoining : The two consonants sit side by side, but the second consonant has a special shape.
   3. Ligation : Create a fusion of the letter shapes, where it may be difficult to identify one or more of the components.
   4. The letter RA has its own idiosyncratic way of combining with other consonants, whether it precedes or follows them.
2. Show a visible virama below the non-final consonants in the cluster.
3. Use the anusvara.

See also finals and clength.

Stacking

The overwhelming majority of conjuncts in Odia are achieved by subjoining a reduced form of the non-initial consonant below the initial.

ɦnɔ

ɭpɔ

kkɔ

In most cases the non-initial consonant is just reduced in size, but in some cases the shape is changed, either by removing the circular top line, or in a more fundamental way.

ktɔ

kɖɔ

However, when TA is the initial consonant, it is sometimes the initial that is reduced and subjoined. In other combinations, however, it retains its full form.

tkɔ

tnɔ

RA in clusters

A trailing RA has a fairly regular appearance as a subjoined glyph below the preceding consonant, although that line may join with the preceding letter shape, and therefore cause a slight change to it.

krɔ

However, like many other Indian scripts, 0B30 at the beginning of a cluster is represented idiosyncratically, and appears as a small, superscript glyph over the top right of the following consonant.

rkɔ

Observation: Unlike Devanagari, it appears that the RA doesn't move over a following vowel sign, such as ା [U+0B3E ORIYA VOWEL SIGN AA].

Ligated forms

Certain combinations of consonants form conjuncts by producing a merged glyph one or both of the original letters may be unrecognisable.

d͡ʒɲɔ

ttɔ

kʰjɔ

The following is a list of combinations that produce such an effect. Click on the items to see the component letters.

କ୍ଷ␣ତ୍ତ␣ଦ୍ଧ␣ଦ୍ଭ␣ଙ୍କ␣ଙ୍ଖ␣ଙ୍ଗ␣ଙ୍ଘ␣ଜ୍ଞ␣ଟ୍ଟ␣ତ୍ତ␣ଦ୍ଭ␣ଦ୍ଦ␣ଦ୍ଧ␣ଧ୍ଯ␣ଧ୍ୟ␣ନ୍ଧ␣ନ୍ଦ␣ବ୍ଦ␣ବ୍ବ␣ମ୍ପ␣ମ୍ଫ␣ମ୍ଭ

Conjoined consonants or a visible halanta

Three letters in particular tend not to stack, but sit alongside the initial consonant in the cluster.

td͡ʒɔ

tjɔ

As can be seen above, the conjoined forms for ʤ and j are identical.

The letter NYA also sits alongside the cluster initial, but the halanta may be shown below the initial letter.

tɲɔ

Observation: Noto, Nirmala, and Kangila fonts all show the halanta below the initial consonant in the first example at fig_conjoined, but Oriya MN and Oriya Sangam MN fonts don't show it.

The halanta is also left showing for borrowed words.d,404 The halanta can be made visible by following it with 200C.

Consonant length

Geminated consonants in the middle of a word can be written using 0B03,n,54 eg. ଦୁଃଖ

Consonant sounds to characters

Encoding vowel-signs

The three circumgraphs can be written as a single character, or as two characters. In 2 of those cases, the second character is a lengthening mark.

The single code point per vowel sign is preferred, however the parts are separated in Unicode Normalisation Form D (NFD), and recomposed in Unicode Normalisation Form C (NFC), so both approaches are canonically equivalent.

Whichever approach is used, the vowel signs must be typed and stored after the consonant characters they surround, and in left to right order.

Independent vowels

The approach listed in the table below is not equivalent when the text is normalised, and therefore only the precomposed approach in the left column should be used.u,487

In addition to the problem previously mentioned, combinations on rows 2 and 3 don't have the joining bar and so won't display correctly.

Digits

Odia has its own set of native digits.

The CLDR standard-decimal pattern is #,##,##0.###. The standard-percent pattern is #,##,##0%.c

ASCII digits may also be used.msg

Fractions

Odia also has a number of pre-decimal characters representing fractions.

These are used additively, with larger values appearing before smaller, eg. 0B73 0B75 represents the value 5/16.u,490

Context-based shaping & positioning

Odia text relies on OpenType rules to correctly position glyphs and shape them according to the surrounding text.

One major area where this applies is in the use of conjunct forms for consonant clusters.

See a table of 2-consonant clusters for Oriya.

The following is a selection of other examples of contextual shaping and positioning.

Positioning u in clusters. When a below-base vowel sign occurs with a cluster with a conjoined form it is attached to the larger glyph, rather than to the consonant it actually follows in memory and speech, eg.

Position & shape of i. After a certain consonant glyphs, in some fonts, the vowel sign for i appears in a different position and with a different shape. The first example in the table below shows the typical shape.

Other glyph variants. Nakanishi lists a number of alternative shapes for glyphs.

Explicit shaping controls

200C (ZWNJ) can be used to force the production of a visible virama, rather than a conjunct form.

200D (ZWJ) is used to produce a ligated version of OM (see symbols).

Word boundaries

Words are separated by spaces.

Hyphens may be used to separate parts of a compound word,msg,40 eg. ଡ୍ରପ୍-ଡାଉନ୍

Graphemes

Conjuncts

Conjuncts and any dependent combining characters should never be split.

This creates a problem when dealing with Unicode grapheme clusters, because they stop after reaching a virama. So conjuncts usually contain multiple grapheme clusters. This produces incorrect segmentation as seen on the right in fig_grapheme_conjunct. Applications need to tailor the grapheme cluster rules to avoid splitting conjuncts.

show composition

ସାଙ୍ଗେ

Unfortunately, this is harder than it seems, because whether a conjunct is formed or not usually depends on the capabilities of the font – it cannot be determined solely by looking at the code points in memory. If a font doesn't contain the glyphs to create a conjunct it will render the consonant cluster with a visible virama. In that case, the grapheme cluster approach is appropriate.

Phrase & section boundaries

Odia uses a combination of ASCII and native punctuation.

0964 and 0965 are from the Unicode Devanagari block. Odia uses a space before these punctuation marks, which avoids confusion with 0B3E, eg.

… ଲୋପ ପାଇଗଲା ।

Quotations & citations

Odia texts typically use quotation marks around quotations. Of course, due to keyboard design, quotations may also be surrounded by ASCII double and single quote marks.

Single quotation marks are used for quotations within quotations.

Abbreviation, ellipsis & repetition

Other inline features

Line breaking & hyphenation

Lines are mostly broken at inter-word spaces.

Like most writing systems, certain characters are expected not to start or end a line. For example, periods and commas shouldn't start a line, and opening parentheses shouldn't end a line.

Show (default) line-breaking properties for characters in the modern Odia orthography.

Baselines, line height, etc.

tbd

Odia uses the so-called 'alphabetic' baseline, which is the same as for Latin and many other scripts.

Counters, lists, etc.

You can experiment with counter styles using the Counter styles converter. Patterns for using these styles in CSS can be found in Ready-made Counter Styles, and we use the names of those patterns here to refer to the various styles.