Title :
Performance of Marathi language TTS synthesis based on perceptual test and spectrogram analysis
Author :
Bormane, Dattatraya S. ; Shirbahadurkar, S.D. ; Shiurka, U.D.
Author_Institution :
JSPM´s, Rajarshi Shahu Coll. of Eng., Pune, India
Abstract :
This paper describes the work based on concatenative text-tospeech synthesis system. It discusses a few perceptual and spectrogram experiments conducted on Marathi Voices (Spoken in Maharashtra, India). Marathi speech synthesizer is developed using different choice of units: words, phonemes as a database. We have synthesized the Marathi text and conducted the perceptual tests, as a result, (1) 74% of speech synthesized by the proposed method was preferred to that by the conventional method, (2) the mean opinion score (MOS) was 3.94 in a five-point MOS test, and 87% of the synthesized speech had the same naturalness as natural speech w.r.t. 40 samples taken from various slot of databases (3) Histogram for various speech databases shows the effectiveness of the proposed method. (4) Spectrogram analysis of various words concatenated with phonemes, syllables as a unit.
Keywords :
natural language processing; speech processing; speech synthesis; Marathi Voices; Marathi language TTS synthesis; concatenative text-tospeech synthesis system; database; five-point MOS test; histogram; mean opinion score; perceptual test; phonemes; spectrogram analysis; speech synthesizer; words; Concatenated codes; Databases; Histograms; Natural languages; Performance analysis; Spectrogram; Speech analysis; Speech synthesis; Synthesizers; Testing; Speech synthesis; concatenation; histogram; unit size;
Conference_Titel :
Computer and Automation Engineering (ICCAE), 2010 The 2nd International Conference on
Conference_Location :
Singapore
Print_ISBN :
978-1-4244-5585-0
Electronic_ISBN :
978-1-4244-5586-7
DOI :
10.1109/ICCAE.2010.5451850
