Rule Based Speech Synthesis by Cepstral Method for Standard Bangla

نویسندگان

  • Shahina HAQUE
  • Tomio TAKARA
چکیده

In the first phase of this paper, we describe the construction of a Bangla speech synthesizer. In the second phase, we discuss our work on Bangla nasal vowel. Nasality is one of the distinctive characteristic of Bangla phonemes. We discuss methods employed for transforming Bangla oral vowel to the corresponding nasal vowel counterpart and its application to our speech synthesizer. The perceptual evaluation result of the system at present, oral vowel 100%, nasal vowel 90%, oral-nasal detection 100%, initial consonant 87% and final consonant is 82%, 86% for non-linear transformation and 91 % for sine curve model.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Performance Evaluation of Bangla Word Recognition Using Different Acoustic Features

This paper describes a medium size Bangla speech corpus preparation and the comparison of the performances of different acoustic features for Bangla word recognition. A small number of speakers are use for most of the Bangla automatic speech recognition (ASR) system, but 40 speakers selected from a wide area of Bangladesh, where Bangla is used as a native language, are involved here. In the exp...

متن کامل

A study on the pitch pattern of a singing voice synthesis system based on the cepstral method

We synthesize singing voice by rule based on cepstral method. Higher accuracy of analysis and synthesis is required to synthesize singing voice, comparing to rule-based speech synthesis. In this paper, we propose a method of analysis and synthesis with high accuracy. Also, we express pitch patterns minutely by curves that close to natural pitch by using this method. We apply Fujisaki model and ...

متن کامل

Local Feature or Mel Frequency Cepstral Coefficients - Which One Is Better for MLN-Based Bangla Speech Recognition?

This paper discusses the dominancy of local features (LFs), as input to the multilayer neural network (MLN), extracted from a Bangla input speech over mel frequency cepstral coefficients (MFCCs). Here, LF-based method comprises three stages: (i) LF extraction from input speech, (ii) phoneme probabilities extraction using MLN from LF and (iii) the hidden Markov model (HMM) based classifier to ob...

متن کامل

Text Normalization System for Bangla

This paper describes a process of text normalization system for the Bangla language (exonym: Bengali) by identifying the semiotic classes from Bangla text corpus. After identifying the semiotic classes, a set of rules was written for tokenization and verbalization. This study is important for Text-ToSpeech (TTS) system and as well as for creating a language model used in speech recognition.

متن کامل

Epoch synchronous non-overlap-add (ESNOLA) method-based concatenative speech synthesis system for Bangla

In the last decade there has been a shift towards development of speech synthesizer using concatenative synthesis technique instead of parametric synthesis. There are a number of different methodologies for concatenative synthesis like TDPSOLA, PSOLA, and MBROLA. This paper, describes a concatenative speech synthesis system based on Epoch Synchronous Non Over Lapp Add (ESNOLA) technique, for st...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2004