Sandhi is the set of euphonic-combination rules inherited from Sanskrit by which sounds at the junction of words or morphemes merge or change. For example अ plus आ combines to आ, and इ plus अ yields य. Hindi keeps many of these in tatsama (Sanskrit-derived) vocabulary.

What are the main categories of sandhi?

The three classical categories are swar sandhi (vowel junctions), vyanjan sandhi (consonant junctions), and visarga sandhi (changes affecting the visarga ः). This tool flags candidate sites for each by examining the characters on either side of a junction.

Why are these only candidates?

Whether sandhi actually applies depends on word origin and meaning, which a surface scan cannot fully determine. Tadbhava and foreign-origin words may share the same letter sequences without undergoing sandhi, so each flag should be confirmed by an editor.

Does it split or join words for me?

No. The tool only highlights where sandhi rules commonly operate and names the relevant category. It does not perform sandhi-viccheda (splitting) or sandhi formation automatically, since both require lexical knowledge of the underlying words.

Hindi Sandhi Junction Identifier

Email me this result

Get this tool's output sent to your inbox, plus one useful tool a week. No spam, unsubscribe any time.

The Hindi Sandhi Junction Identifier highlights places in Devanagari text where Sanskrit-derived sandhi rules commonly apply. It is an editing aid for spotting vowel coalescence, visarga changes, and consonant junctions in tatsama vocabulary.

How it works

The scanner walks the text and inspects the characters on either side of each junction — both between words and at internal morpheme seams — testing them against the three classical categories:

swar sandhi    vowel meets vowel        अ+आ → आ,  इ+अ → य,  उ+अ → व
visarga sandhi visarga ः before sound    निः+चल → निश्चल,  मनः+रथ → मनोरथ
vyanjan sandhi consonant junctions      त्+च → च्च,  म्+anusvara before stop

For each match the tool records the position and the rule category. Junctions at word boundaries are checked by combining the final sound of one word with the initial sound of the next; internal positions are checked against known clustering patterns.

Tips and notes

Treat every highlight as a candidate, not a certainty. Sandhi only operates in Sanskrit-derived (tatsama) material, so a sequence such as इ followed by अ may be a genuine swar-sandhi site in one word and an accidental adjacency in a borrowed word. Use the named category to decide which rule to apply, and confirm against the meaning and origin of the words involved. The tool deliberately does not rewrite your text, because both forming and splitting sandhi require knowing the underlying lexemes.