How to Type Special Characters — Halant, Nukta, Chandrabindu, Visarga

Hindi has several characters that don't map neatly to any English letter. These are the ones that cause "how do I type THIS?" moments for almost everyone — even native Hindi speakers new to digital typing.

The Characters That Cause Trouble

Character	Name	Looks Like	What It Does
्	Halant/Virama	Small diagonal stroke below	Kills the inherent "a" vowel
़	Nukta	Dot below	Creates borrowed sounds (ज़, फ़, etc.)
ँ	Chandrabindu	Moon-dot above	Nasal vowel sound
ं	Anusvara	Dot above	Nasal consonant sound
ः	Visarga	Two dots to the right	Aspirated breath after vowel
ॅ	Chandrakala	Half-moon above	Short English "a" sound (as in "cat")
ऽ	Avagraha	Looks like S	Indicates dropped "a" in Sanskrit

Halant (्) — The Vowel Killer

Every Hindi consonant has an inherent "a" sound. क is "ka", not just "k". The halant removes that "a" — turning क into क्.

When you need it:

• Conjunct consonants: क् + र = क्र (kra)
• Word-final consonants without vowel: महात्मन् (the final न needs halant)

How to type it:

Method	Input
Phonetic transliteration	Usually automatic — type kr and the tool creates क्र
InScript keyboard	d key (halant key on InScript layout)
Unicode direct	U+094D

Phonetic transliteration handles halant automatically in almost all cases. When you type namaste, the tool knows "st" means स् + त (halant on स, then त). You rarely need to manually insert a halant.

The exception: when a tool misinterprets your intent. If you type santa and get सन्ता instead of सांता, you may need to add a space or vowel to guide the tool. On TranslitHub, the suggestion dropdown usually shows both options.

Nukta (़) — The Borrowed Sound Dot

Hindi borrowed several sounds from Persian/Arabic that don't exist in native Sanskrit-derived words. The nukta (a dot below the consonant) marks these borrowed sounds.

Without Nukta	With Nukta	Sound Change
क (ka)	क़ (qa)	Hard "q" as in Qutub
ख (kha)	ख़ (kha)	Rough "kh" as in Khan
ग (ga)	ग़ (gha)	Soft gargling "gh"
ज (ja)	ज़ (za)	"z" as in ज़िन्दगी (zindagi)
फ (pha)	फ़ (fa)	"f" as in फ़ोन (phone)
ड (Da)	ड़ (Dra)	Flapped "r" sound
ढ (Dha)	ढ़ (Drha)	Aspirated flap

How to type it:

Method	Input
Phonetic	z → ज़, f → फ़, q → क़
InScript	Type the base consonant, then the nukta key (] key)
Mobile (Gboard)	Long-press the base consonant → select nukta variant

Common mistake: Many people skip the nukta and type ज instead of ज़. In casual writing, this is common and generally understood. In formal writing, exam papers, or published content, the nukta matters.

Chandrabindu (ँ) — The Nasal Moon

The chandrabindu nasalizes the vowel it sits on. It's different from the anusvara (ं) — the chandrabindu affects the vowel sound itself, while the anusvara represents a nasal consonant.

Examples:

• हँसना (to laugh) — the अ vowel in ह is nasalized
• माँ (mother) — the आ vowel is nasalized
• गाँव (village) — nasalized आ
• चाँद (moon) — nasalized आ

How to type it:

Method	Input
Phonetic	maa~ or maan (varies by tool) — many tools use ~ or detect from context
InScript	Shift + Chandrabindu key

The anusvara shortcut: In modern Hindi, many writers use anusvara (ं) where chandrabindu (ँ) is technically correct. "माँ" and "मां" are both common. Strictly, chandrabindu is correct when the vowel matra extends above the headline (as in आ, ई, ऊ, ए, ऐ, ओ, औ), and anusvara is used otherwise.

In practice, using anusvara everywhere is accepted in all but the most formal contexts.

Anusvara (ं) — The Nasal Dot

The anusvara adds a nasal sound — either nasalizing the vowel or representing a nasal consonant (ङ, ञ, ण, न, म) before another consonant.

Examples:

• हिंदी (Hindi) — the anusvara represents the nasal "n" before "d"
• संस्कृत (Sanskrit) — nasal "n" before "s"
• अंगूर (grapes) — nasal "ng" before "g"

How to type it:

Method	Input
Phonetic	Type n before a consonant: hindi → हिंदी (automatic)
Explicit	Some tools use .n or M for explicit anusvara
InScript	x key (standard position)

Most transliteration tools handle anusvara automatically. When you type hindi, the tool knows the "n" before "d" should be an anusvara on the इ matra.

Visarga (ः) — The Breath Mark

The visarga adds an aspirated "h" sound after a vowel. It's rare in modern Hindi but common in Sanskrit-derived formal words.

Examples:

• दुःख (sorrow) — duHkh
• अतः (therefore) — ataH
• प्रातः (morning) — praataH
• नमः (salutation) — namaH (as in ॐ नमः शिवाय)

How to type it:

Method	Input
Phonetic	H (capital H) at end of syllable: duHkh → दुःख
InScript	Shift + corresponding key

Common confusion: The visarga (ः) is NOT the same as typing a colon (:). The visarga is a Devanagari character with its own Unicode codepoint (U+0903). Using an English colon will look similar but break text processing, search, and sorting.

Avagraha (ऽ) — The Dropped A

The avagraha marks a dropped "a" sound, mainly in Sanskrit sandhi. You'll see it in:

• मनोऽभिलाषा (mano'bhilaashaa)


• देवोऽस्ति (devo'sti)

In modern Hindi, the avagraha is rare. You'll mainly encounter it in religious texts, classical poetry, and formal Sanskrit.

How to type: Most phonetic tools map the apostrophe (') to avagraha in Hindi mode.

Script-Specific Special Characters

Tamil: Aytham (ஃ)

The aytham (visarga equivalent in Tamil) is typed as aH or q in most Tamil transliteration tools.

Bengali: Hasanta and Chandrabindu

Bengali uses the same concepts (hasanta = halant, chandrabindu) but with slightly different rendering. Bengali chandrabindu (ঁ) sits differently than Devanagari chandrabindu.

Gujarati: No Shirorekha

Gujarati lacks the headline (shirorekha) that Devanagari has, so some characters look similar but are distinct Unicode points. Don't copy-paste Devanagari characters into Gujarati text.

Urdu: Hamza and Ain

Urdu has characters from Arabic that Indian scripts don't: hamza (ء), ain (ع). These require specific input sequences in transliteration tools.

Quick Reference Card

Save or print this for reference:

Character	Name	Phonetic Input	When to Use
्	Halant	Automatic (between consonants)	Conjuncts, pure consonants
़	Nukta	z, f, q (on base consonant)	Urdu/Persian borrowed sounds
ँ	Chandrabindu	~ or contextual	Nasalized vowels (माँ, गाँव)
ं	Anusvara	n (before consonant)	Nasal sounds (हिंदी, संगीत)
ः	Visarga	H (capital)	Sanskrit words (दुःख, अतः)
ॐ	Om	om	Religious context
॥	Double Danda	Verse endings
।	Danda (purna viram)	. or	Hindi full stop

For everyday Hindi typing, you'll use anusvara and chandrabindu most frequently. The halant works automatically in transliteration tools. Nukta matters for words with ज़ and फ़. Visarga is rare outside formal or religious text.

Tools like TranslitHub handle most of these automatically — the suggestion dropdown shows variants when ambiguity exists, so you can pick the correct one without memorizing obscure input codes.