Towards automatic phonetic segmentation for TTS

Rendel, Asaf; Sorin, Alexander; Hoory, Ron; Breen, Andrew P.

doi:10.1109/icassp.2012.6288926

Cited by 9 publications

(3 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We proposed a simple iterative procedure that helps to improve the segmentation accuracy so that it is usable in a concatenative TTS system, for which precise segmentation is very important. 62 The procedure described has recently been utilized in our TTS system 61,63 when adding new voices in new languages.…”

Section: Research Objectivementioning

confidence: 99%

Using LSTM neural networks for cross‐lingual phonetic speech segmentation with an iterative correction procedure

Hanzlíček,

Matoušek,

Vít

2023

Computational Intelligence

View full text Add to dashboard Cite

This article describes experiments on speech segmentation using long short‐term memory recurrent neural networks. The main part of the paper deals with multi‐lingual and cross‐lingual segmentation, that is, it is performed on a language different from the one on which the model was trained. The experimental data involves large Czech, English, German, and Russian speech corpora designated for speech synthesis. For optimal multi‐lingual modeling, a compact phonetic alphabet was proposed by sharing and clustering phones of particular languages. Many experiments were performed exploring various experimental conditions and data combinations. We proposed a simple procedure that iteratively adapts the inaccurate default model to the new voice/language. The segmentation accuracy was evaluated by comparison with reference segmentation created by a well‐tuned hidden Markov model‐based framework with additional manual corrections. The resulting segmentation was also employed in a unit selection text‐to‐speech system. The generated speech quality was compared with the reference segmentation by a preference listening test.

show abstract

Section: Research Objectivementioning

confidence: 99%

Using LSTM neural networks for cross‐lingual phonetic speech segmentation with an iterative correction procedure

Hanzlíček,

Matoušek,

Vít

2023

Computational Intelligence

View full text Add to dashboard Cite

show abstract

“…As the automatic phonetic segmentation accuracy has attracted researchers for many years, a number of HMM-based force alignment framework refinements were proposed (see, e.g., [1][2][3][4][5]). On the other hand, the origin of gross segmentation errors and a way to fix them has not been researched so much.…”

Section: Introductionmentioning

confidence: 99%

Annotation errors detection in TTS corpora

Matoušek

Tihelka

2013

Interspeech 2013

View full text Add to dashboard Cite

We investigate the problem of automatic detection of annotation errors in single-speaker read-speech corpora used for textto-speech (TTS) synthesis. Various word-level feature sets were used, and the performance of several detection methods based on support vector machines, extremely randomized trees, knearest neighbors, and the performance of novelty and outlier detection are evaluated. We show that both word-and utterancelevel annotation error detections perform very well with both high precision and recall scores and with F 1 measure being almost 90%, or 97%, respectively.

show abstract

“…TIMIT-korpuse automaatsegmentimisel on saadud tulemuseks 79,5% ja 92,8% häälikupiiridest vastavalt 10 ms ja 20 ms sees (Hosom 2009). Üks hilisem uuring tutvustab automaatsegmentimise meetodit, mis on andnud sama korpuse puhul tulemuseks vastavalt 84,6% ja 95,4% (Rendel & Sorin et al 2012).…”

unclassified

Automaatse segmentimise hindamine

Meister

2017

View full text Add to dashboard Cite

Töös hinnatakse kahe eestikeelse kõne automaatsegmentimise programmiga leitud hääliku-ja sõnapiiride erinevust võrreldes käsitsi määratud piiridega ning uuritakse, kui palju erinevad käsitsi ja automaatselt saadud märgendustest mõõdetud segmendikestused. Uuringus kasutatakse 14 keelejuhi kõnenäiteid (kokku 208 lauset), neist neli eesti emakeelega (L1) täiskasvanut, kuus L1 last ning neli eesti keelt võõrkeelena (L2) kõnelevat täiskasvanut. Tulemused näitavad, et mõlemad automaatsüsteemid annavad paremaid tulemusi L1 täiskasvanute kõne puhul võrreldes lastekõne ja L2 kõnega. Automaatsegmenditud materjalist mõõdetud segmendikestused on lähedased käsitsi segmenditud kõnest leitud kestustele.

show abstract

Towards automatic phonetic segmentation for TTS

Cited by 9 publications

References 8 publications

Using LSTM neural networks for cross‐lingual phonetic speech segmentation with an iterative correction procedure

Using LSTM neural networks for cross‐lingual phonetic speech segmentation with an iterative correction procedure

Annotation errors detection in TTS corpora

Automaatse segmentimise hindamine

Contact Info

Product

Resources

About