"Ü", or "ü", is a glyph which represents either a letter from several extended Latin alphabets, the letter U with umlaut, or a letter U with diaeresis.
The letter Ü occurs in the Estonian, Hungarian, and Turkish alphabets, where it represents a front rounded high vowel (/y/).
This same letter appears in the Chinese romanizations pinyin, Wade-Giles, and the German-based Lessing-Othmer , where it represents the same sound: the vowel of 玉 (jade) and 雨 (rain). Pinyin uses Ü only when ambiguity could arise with similarily romanized words containing a U, whereas Wade-Giles and Lessing use Ü in all situations.
A similar glyph, U with umlaut, appears in the German alphabet. It represents the umlauted form of u, which results in the same sound as the letter Ü mentioned in the previous section: /y/. The letter is collated together with U. In languages which have adopted German names or spellings, such as Swedish and Dutch, the letter also occurs. It is however not a part of these languages' alphabets.
In other languages that do not have the letter as part of the regular alphabet or in limited character sets such as ASCII, U-umlaut is frequently replaced with the two-letter combination "ue".
U with diaeresis occurs in several languages which use diaereses.
In Portuguese, Castillian and Catalan Ü performs a special rôle. U is normally silent in these languages, and Ü is used to indicate the vowel is to be pronounced in a position where it would normally be silent, specifically between a "G" and an "E" or "I" and between a "Q" and an "E" or "I" in Portuguese and Catalan. Its sound remains a normal U as used in these languages.
Since in Spanish the pronunciation of G ordinarily changes from /g/ to /x/ (same as Spanish "J") when followed by I or E, a silent U is added after the G when it is necessary to preserve the /g/ sound. For example, guerra (war) is pronounced /geRa/, not */xeRa/. However, in cases where the U should not be silent, a diaeresis is added. For example, lingüística (linguistics) is pronounced /lingwistika/, not */lingistika/. It is used similarly in Portuguese (only in Brazil) and Catalan.
Other languages which use diaereses, such as French or Dutch, may contain other occurences of Ü. Catalan also contains words with Ü which are not subject to the pronunciation changing rules given above. Since some of these languages also contain U-umlaut, which looks identical to U-diaeresis, words may be inadvertently mispronounced.
Historically the unique letter Ü and U-diaeresis were written as a U with two dots above the letter. U-umlaut was written as a U with a small e written above: this minute e degenerated to two vertical bars in medieval handwritings. In most later handwritings these bars in turn nearly became dots.
In modern typography there was insufficient space on typewriters and later computer keyboards to allow for both a U-with-dots (also representing Ü) and a U-with-bars. Since they looked near-identical the two glyphs were combined, which was also done in computer character encodings such as ISO 8859-1. As a result there was no way to differentiate between the three different characters. While Unicode theoretically provides a solution, this is almost never used.
The HTML entity for Ü is Ü. For ü, it is ü (Mnemonic for "U umlaut").