Assessing the quality of TTS audio in the LARA learning-by-reading
                        platform

Akhlaghi, Elham; Bączkowska, Anna; Berthelsen, Harald; Bédi, Branislav; Chua, Cathy; Cucchiarini, Catia; Habibi, Hanieh; Horváthová, Ivana; Hvalsøe, Pernille; Lotz, Roy; Maizonniaux, Christèle; Chiaráin, Neasa Ní; Rayner, Manny; Tsourakis, Nikos; Yao, Chunlin

doi:10.14705/rpnet.2021.54.1299

Cited by 3 publications

(4 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One expects TTS audio to be much quicker to produce, but of lower quality: the goal was to obtain quantitative and qualitative data exploring the issues. The results were presented at EUROCALL 2021 (Akhlaghi et al, 2021). To our surprise, TTS audio was in fact rated equal to or better than human audio in three of the ten languages.…”

Section: Reading Assistance Through Integrated Ttsmentioning

confidence: 94%

Overview of LARA: A Learning and Reading Assistant

Akhlaghi¹,

Bédi²,

Butterweck³

et al. 2019

8th ISCA Workshop on Speech and Language Technology in Education (SLaTE 2019)

View full text Add to dashboard Cite

We present an overview of LARA, the Learning And Reading Assistant, an open source platform for easy creation and use of multimedia annotated texts designed to support the improvement of reading skills. The paper is divided into three parts. In the first, we give a brief summary of LARA's processing. In the second, we describe some generic functionality specially relevant for reading assistance: support for phonetically annotated texts, support for image-based texts, and integrated production of text-to-speech (TTS) generated audio. In the third, we outline some of the larger projects so far carried out with LARA, involving development of content for learning second and foreign (L2) languages such as Icelandic, Farsi, Irish, Old Norse and the Australian Aboriginal language Barngarla, where the issues involved overlap with those that arise when trying to help students improve first-language (L1) reading skills. All software and almost all content is freely available.

show abstract

Section: Reading Assistance Through Integrated Ttsmentioning

confidence: 94%

Overview of LARA: A Learning and Reading Assistant

Akhlaghi¹,

Bédi²,

Butterweck³

et al. 2019

8th ISCA Workshop on Speech and Language Technology in Education (SLaTE 2019)

View full text Add to dashboard Cite

show abstract

“…Audio was made available in human-recorded form for sentences and in TTS voice for words. TTS audio was produced by the ReadSpeaker engine, which at a word level gives quality judged comparable to human audio (Akhlaghi et al 2021).…”

Section: Process/methodsmentioning

confidence: 99%

“…We now briefly describe how these annotations are created. Full details can be found in the online documentation (Rayner et al, 2021).…”

Section: The Lara Platformmentioning

confidence: 99%

See 1 more Smart Citation

Using the LARA platform to crowdsource a multilingual, multimodal Little Prince

Akhlaghi

Bączkowska²,

Bédi³

et al. 2022

beyond

View full text Add to dashboard Cite

We describe an ongoing project, in which an informally organised international consortium is using the open source LARA platform to create multimodal annotated editions of Antoine de Saint-Exupéry’s Le petit prince in multiple languages, so far French, English, Italian, Icelandic, Irish, Japanese, Polish, Farsi and Mandarin. LARA versions of the book include integrated audio and translations and an automatically generated lemma-based concordance, and are freely available online. We describe the methods used to construct the various versions. In some cases, work for a given language was simply divided by type, typically with one person adding translations and another recording audio. In other languages, we experimented with crowdsourcing methods, splitting the text into chapter-sized units and using the LARA platform to distribute these to multiple annotators, then combining the results at the end. Finally, we report an initial classroom study, where the French version was used by intermediate-level Australian students of French.

show abstract

Assessing the quality of TTS audio in the LARA learning-by-reading platform

Akhlaghi

Bączkowska

Berthelsen

et al. 2021

CALL and Professionalisation: Short Papers From EUROCALL 2021

View full text Add to dashboard Cite

A popular idea in Computer Assisted Language Learning (CALL) is to use multimodal annotated texts, with annotations typically including embedded audio and translations, to support L2 learning through reading. An important question is how to create the audio, which can be done either through human recording or by a Text-To-Speech (TTS) synthesis engine. We may reasonably expect TTS to be quicker and easier, but humans to be of higher quality. Here, we report a study using the open-source LARA platform and ten languages. Samples of LARA audio totaling about three and a half minutes were provided for each language in both human and TTS form; subjects used a web form to compare different versions of the same item and rate the voices as a whole. Although human voice was more often preferred, TTS achieved higher ratings in some languages and was close in others.

show abstract

Assessing the quality of TTS audio in the LARA learning-by-reading platform

Cited by 3 publications

References 1 publication

Overview of LARA: A Learning and Reading Assistant

Overview of LARA: A Learning and Reading Assistant

Using the LARA platform to crowdsource a multilingual, multimodal Little Prince

Assessing the quality of TTS audio in the LARA learning-by-reading platform

Contact Info

Product

Resources

About