论文部分内容阅读
In this paper,we introduce a speaker-dependent isolated word recognizer which isdedicated for Chinese character input.The method presented here offers an effective solution tothe large-vocabulary recognition problem by carrying out recognition hierachically.Thevocabulary consists of 800 to 1000 words.The average recognition rate is 90% whenmonosyllable words takes up one third of the vocabulary.Recognition rate can reach 95% byselecting from the top 20 candidates.
In this paper, we introduce a speaker-dependent isolated word recognizer which isdedicated for Chinese character input.The method presented offers an effective solution for tothe large-vocabulary recognition problem by carrying out recognition hierachically.Thevocabulary consists of 800 to 1000 words. Average recognition rate is 90% whenmonosyllable words takes up one third of the vocabulary. Recognition rate can reach 95% byselecting from the top 20 candidates.