Building a better Indian English voice using "more data"
نویسندگان
چکیده
Q: (Under given circumstances) Can we improve speech synthesis by using “more” data ? If so, How much more data is needed ? What kind of data is useful ? Circumstances we are interested in: Improve an existing Indian English Voice (CMU ARCTIC KSP) “More” data that we had for these experiments New Indian English Pronunciation Lexicon (~3500 new words with Indian English Pronunciations) New Indian English Acoustic Data (besides the CMU ARCTIC KSP acoustic data) Our Approach: Use additional data to create new
منابع مشابه
Factors Contributing to the Use of L1 in English Classrooms: Listening to the Voice of Teachers and Students in Iranian Institutes
By the growing emphasis on communicative language skills, the attention toward using first language in second language classrooms has been raised. The primary aim of this study was to find out teachers’ and students’ perceptions toward the utilization of L1 in English classrooms in Institutes. In order to do this objective, the researcher used a mixed methods research design. Two gr...
متن کاملVoice Analysis in English and Persian Persuasive Texts: Pedagogical implications in focus
The main purpose of this study is to investigate how voice is realized by Iranian EFL learners in persuasive English and Persian text types. This discourse-related notion is a required criterion for writing acceptable English. However, L2 learners from cultures other than English might face problems in realizing it, or even ignore it all through their writing. In this connection, the present st...
متن کاملVoice Analysis in English and Persian Persuasive Texts: Pedagogical implications in focus
The main purpose of this study is to investigate how voice is realized by Iranian EFL learners in persuasive English and Persian text types. This discourse-related notion is a required criterion for writing acceptable English. However, L2 learners from cultures other than English might face problems in realizing it, or even ignore it all through their writing. In this connection, the present st...
متن کاملA Pragmatic Solution to an Indian Accented English Speech Synthesizer Using Residual Excited Linear Predictive Coded Voice
This paper elucidates a practical solution to an Indian accented English text to speech synthesizing system. The paper covers the complete procedure to generate the speech signal of the text, in Indian accented voice. The technique described considers the various prosodic features that need to be incorporated into the synthesized speech to make it appear natural and in the way an Indian speaks....
متن کاملVoice in Short Argumentative Texts Written by Undergraduate Learners of English
The present study explored the intensity level of authorial voice in relation to the quality of argumentative writing. 42 undergraduate learners of English as a foreign language (36 girls and 6 boys) spent 45 minutes to individually complete in-class position-taking writing tasks for three weeks. Their overall academic writing quality scores assigned based on portfolio assessment were studied i...
متن کامل