▎ 摘　　要

NOVELTY - The method involves identifying (202) a word or phrase as a named entity using one or more processors. A language of origin is identified (204) that is associated with the named entity. The named entity is transliterated (206) to a script that is associated with the language of origin. The transliterated script is passed (208) to the TTS system if the TTS system is operating in the language of origin. A phoneme sequence is generated (210) in the language of origin that uses a graphene to phoneme (G2P) converter if the TTS system is not operating in the language of origin. USE - Computer-implemented method configured for use in text-to-speech (TTS) system (claimed). ADVANTAGE - The machine transliteration (MT) is the process of automatically converting the script of a word from a source language (e.g. English) to a target language (e.g. Hindi). The advancement in data mining technology is combined with transliteration technology to provide a novel solution to improve the pronunciation of names. The TTS system or automatic speech recognition (ASR) system a dictionary can be required to generate accurate pronunciations or to recognize words. The TTS can pronounce names that are native to its language much more accurately than names that are foreign to it. The pronunciation process can be configured to accurately generate pronunciation of a foreign name in the TTS's phoneme set using a number of approaches and to locate the best map for the phoneme sequence from Hindi so that the resulting system sounds close to Hindi system. DETAILED DESCRIPTION - An INDEPENDENT CLAIM is included for a non-transitory computer-readable storage medium having stored instructions. DESCRIPTION OF DRAWING(S) - The drawing shows a flowchart of a pronunciation process. Step for identifying word phrase (202) Step for identifying language of origin (204) Step for transliterating to script (206) Step for passing transliterated script (208) Step for generating phoneme sequence (210)