---
title: "AusTalk: an audio-visual corpus of Australian English (2014)"
description: "The paper presents AusTalk, a large audio-visual corpus of Australian English (AusE) built within the… 16 references, cited by 2 papers here."
authors: ["Dominique Estival", "Steve Cassidy", "Felicity Cox", "Denis Burnham"]
year: 2014
venue: "Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14)"
source: "https://aclanthology.org/L14-1432/"
licence: "CC BY-NC 4.0 (ELRA, LREC proceedings)"
licence_url: "https://creativecommons.org/licenses/by-nc/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/estival-2014-austalk-audio-visual-corpus-australian/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/estival-2014-austalk-audio-visual-corpus-australian/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2014](https://md.research.elamur.ai/free/en/papers/2014/page/3/) / AusTalk: an audio-visual corpus of Australian English

# AusTalk: an audio-visual corpus of Australian English

**Authors:** [Dominique Estival](https://md.research.elamur.ai/free/en/authors/dominique-estival/), Steve Cassidy, Felicity Cox and Denis Burnham · **Year:** 2014 · **Venue:** [Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14)](https://md.research.elamur.ai/free/en/venues/lrec/) ([LREC 2014](https://md.research.elamur.ai/free/en/venues/lrec/2014/)) · **Publisher:** [European Language Resources Association (ELRA)](https://md.research.elamur.ai/free/en/publishers/european-language-resources-association-elra/) · **Licence:** [CC BY-NC 4.0 (ELRA, LREC proceedings)](https://creativecommons.org/licenses/by-nc/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/L14-1432/) · [PDF](https://aclanthology.org/L14-1432.pdf) · [OpenAlex](https://openalex.org/W2251881015)

**Citations:** 21 citations (OpenAlex, 2026-10-14) · [2](https://md.research.elamur.ai/free/en/papers/estival-2014-austalk-audio-visual-corpus-australian/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
The paper presents AusTalk, a large audio-visual corpus of Australian English (AusE) built within the Big ASC project, a collaboration of 11 institutions. The project had two aims: to set up uniform recording infrastructure across Australia and to collect a corpus capturing social, regional and ethno-cultural variation in AusE. Twelve identical hardware/software "Black Boxes" ran a fixed, largely automated procedure, the Standard Speech Collection Protocol (SSCP). This made field recordings at many sites comparable to studio data. Each speaker attends three one-hour sessions of read and spontaneous tasks. Quality was kept up through central training of recording assistants, test recordings at each site, an automatic checking tool, and published per-item quality grades. Fully annotating all the material was not affordable, so the team combined a manually transcribed subset with automatic phonemic alignment through an AusE-adapted MAUS. The corpus is now accessible through the Alveo virtual lab and already supports work in phonetics, speaker perception and audio-visual speech recognition.

## Key points
- Target size is 1000 speakers and close to 3000 hours. By March 2014, almost 800 speakers had been recorded at 13 sites, giving over 2000 uploaded sessions and 22TB of data.
- Each speaker records three sessions spaced a week or more apart, covering nine components in all, of which four involve reading aloud and five involve unscripted speech:
  - read tasks: Words, Digits, Sentences, Story Reading;
  - spontaneous tasks: Yes/No answers, Story Re-telling, Interview, Map Task, Conversation.
  - Capture uses five microphones and two stereo cameras.
- Speakers qualify only if they completed primary and secondary school in Australia; birth in Australia is not required.
- The SSCP-QC tool checks file counts, audio silence/loudness, and video frame skipping/brightness. Final quality grades are A, B, C or D and are stored with the metadata. Annotations live in the DADA store.
- MAUS was trained for AusE on 59 manually transcribed sentences from each of 100 speakers and produces Praat textgrids. Five speakers received full manual phonemic segmentation as a benchmark. Four speakers from the 1993–95 ANDOSL corpus were re-recorded, providing longitudinal data.

## Interesting ideas
- Quality checks are not used only to discard data: each item keeps a graded status (fine, imperfect, unusable, missing) that users can filter on later.
- A small, carefully hand-annotated subset serves both as training data for an automatic tool and as a fixed quality benchmark for correcting that tool's output.
- The team found that people misread scripts, so the prompt text cannot be treated as ground truth for what was actually said.
- The Map Task was redesigned so its landmarks deliberately elicit specific phonological features, turning a natural task into targeted coverage.
- Pairing story reading with a free retelling of the same story yields matched content in two styles.

## For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

- **I1** (learn): silver sentences agreed by witness parsers could carry a similar A–D grade, so training can weight or exclude data by grade instead of a binary keep/drop.
- **I2** (tree): a compact hand-checked gold set could train repair or conversion rules for the parsers and also measure how far automatic trees drift from it.
- **I3** (learn): any "reference" that comes from a template or prompt (e.g., generated test sentences) should be verified before it is used as gold, which also works as a negative control.
- **I4** (answer): QA items on fables or word problems could be authored to force particular constructions or reasoning steps, so each test checks a known phenomenon.
- **I5** (back): retold versus original texts offer natural paraphrase pairs to test whether text → graph → text keeps the same meaning graph across different wordings.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [Macquarie University](https://md.research.elamur.ai/free/en/universities/macquarie-university/) — University, Sydney, Australia · 42 works · 5,776 citations
- [Western Sydney University](https://md.research.elamur.ai/free/en/universities/western-sydney-university/) — University, Richmond, Australia · 4 works · 63 citations

📚 Shelves (6): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works · [💭 Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/) · 1,552 works · [💭 Experimental and Cognitive Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/experimental-and-cognitive-psychology/) · 870 works

🧭 Topics (2): [English](https://md.research.elamur.ai/free/en/topics/english/) · 2,263 works · [🎙️ Speech & spoken language](https://md.research.elamur.ai/free/en/topics/speech/) · 480 works

## Cite

Cite as: Dominique Estival, Steve Cassidy, Felicity Cox and Denis Burnham. 2014. AusTalk: an audio-visual corpus of Australian English. Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14).

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{estival2014austalk,
  title = {{AusTalk: an audio-visual corpus of Australian English}},
  author = {Dominique Estival, Steve Cassidy, Felicity Cox and Denis Burnham},
  booktitle = {Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14)},
  year = {2014},
  doi = {10.63317/5fmuqesg3xoi},
  url = {https://doi.org/10.63317/5fmuqesg3xoi}
}
```

- [← previous Arabic Spelling Correction using Supervised Learning](https://md.research.elamur.ai/free/en/papers/hassan-2014-arabic-spelling-correction-supervised-learning/)
- [next → Automatic Arabic diacritics restoration based on deep nets](https://md.research.elamur.ai/free/en/papers/sallab-2014-automatic-arabic-diacritics-restoration-based/)
- [↑ back to papers of 2014](https://md.research.elamur.ai/free/en/papers/2014/page/3/)

## References

16 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. ABS. 2012. Cultural Diversity in Australia. [http://www.abs.gov.au/ausstats/abs@.nsf/Lookup/2071.0mai](http://www.abs.gov.au/ausstats/abs@.nsf/Lookup/2071.0mai) n+features902012-2013. Australian Bureau of Statistics. [↗](http://www.abs.gov.au/ausstats/abs@.nsf/Lookup/2071.0mai)
2. Anderson, A.H., M. Bader, E.G. Bard, E. Boyle, G. Doherty, S. Garrod, S. Isard, J. Kowtko, J. McAllister, J. Miller, C. Sotillo, H.S. Thompson, and R. Weinert. 1991. "The HCRC Map Task Corpus." Language and Speech no. 34 (4):351-366. [↗](https://doi.org/10.1177/002383099103400404)
3. Boersma, Paul, and David Weenink. 2001. "Praat, a system for doing phonetics by computer." Glot International no. 5 (9/10):341-345. Burnham, Denis, E. Ambikairajah, Joanne. [find](https://scholar.google.com/scholar?q=Boersma%2C+Paul%2C+and+David+Weenink%2E+2001%2E+%22Praat%2C+a+system+for+doing+phonetics+by+computer%2E%22+Glot+International+no%2E+5+%289%2F10%29%3A341%2D345%2E+Burnham%2C+Denis%2C+E%2E+Ambikairajah%2C+Joanne%2E)
4. Arciuli, Mohammed Bennamoun, C.T. Best, Steven Bird, A.B. Butcher, Steve Cassidy, G. Chetty, F.M. Cox, Anne Cutler, Robert Dale, Julien R. Epps, Janet M. Fletcher, Roland Goecke, David B. Grayden, John T. Hajek, John C. Ingram, Shun Ishihara, Nenagh Kemp, Yuko Kinoshita, T. Kuratate, T.W. Lewis, D.E. Loakes, Mark Onslow, David M. Powers, P. Rose, Roberto Togneri, D. Tran, and Michael Wagner. 2009. A blueprint for a comprehensive Australian English auditory-visual speech corpus. In 2008 HCSNet Workshop on Designing the Australian National Corpus. Sydney: Somerville, MA, USA: Cascadilla Proceedings Project. [find](https://scholar.google.com/scholar?q=Arciuli%2C+Mohammed+Bennamoun%2C+C%2ET%2E+Best%2C+Steven+Bird%2C+A%2EB%2E+Butcher%2C+Steve+Cassidy%2C+G%2E+Chetty%2C+F%2EM%2E+Cox%2C+Anne+Cutler%2C+Robert+Dale%2C+Julien+R%2E+Epps%2C+Janet+M%2E+Fletcher%2C+Roland+Goecke%2C+David+B%2E+Grayden%2C+Joh)
5. Burnham, Denis, Steve Cassidy, Felicity Cox, and Robert Dale. 2009. The Big Australian Speech Corpus: An Audio-Visual Speech Corpus of Australian English Australian Re-search Council Linkage, Infrastructure, Equipment and Facilities Grant. Original edition, LE100100211. [find](https://scholar.google.com/scholar?q=Burnham%2C+Denis%2C+Steve+Cassidy%2C+Felicity+Cox%2C+and+Robert+Dale%2E+2009%2E+The+Big+Australian+Speech+Corpus%3A+An+Audio%2DVisual+Speech+Corpus+of+Australian+English+Australian+Re%2Dsearch+Council+Linkage%2C+Infrastr)
6. Burnham, Denis, Dominique Estival, Peter Bugeia, Peter Sefton, and Steven Cassidy. 2012. Above and Beyond Speech, Language and Music: A Virtual Lab for Human Communication Science (HCS vLab). NeCTAR (National eResearch Collaboration Tools & Resources) Virtual Laboratory. Original edition, VL222. [find](https://scholar.google.com/scholar?q=Burnham%2C+Denis%2C+Dominique+Estival%2C+Peter+Bugeia%2C+Peter+Sefton%2C+and+Steven+Cassidy%2E+2012%2E+Above+and+Beyond+Speech%2C+Language+and+Music%3A+A+Virtual+Lab+for+Human+Communication+Science+%28HCS+vLab%29%2E+NeCTAR+%28)
7. Burnham, Denis, Dominique Estival, Steven Fazio, Felicity Cox, Robert Dale, Jette Viethen, Steve Cassidy, Julien Epps, Roberto Togneri, Yuko Kinoshita, Roland Göcke, Joanne Arciuli, Marc Onslow, Trent Lewis, Andy Butcher, John Hajek, and Michael Wagner. 2011. Building an audio-visual corpus of Australian English: large corpus collection with an economical portable and replicable Black Box. In Interspeech 2011. Florence, Italy. [find](https://scholar.google.com/scholar?q=Burnham%2C+Denis%2C+Dominique+Estival%2C+Steven+Fazio%2C+Felicity+Cox%2C+Robert+Dale%2C+Jette+Viethen%2C+Steve+Cassidy%2C+Julien+Epps%2C+Roberto+Togneri%2C+Yuko+Kinoshita%2C+Roland+G%C3%B6cke%2C+Joanne+Arciuli%2C+Marc+Onslow%2C+Trent)
8. Cassidy, Steve, Dominique Estival, Timothy Jones, Denis Burnham, and Jared Burghold. 2014. The Human Communication Science Virtual Laboratory: A Web Based Repository API. In The 9th edition of the Language Resources and Evaluation Conference (LREC 2014). Reykjavik, Iceland. [find](https://scholar.google.com/scholar?q=Cassidy%2C+Steve%2C+Dominique+Estival%2C+Timothy+Jones%2C+Denis+Burnham%2C+and+Jared+Burghold%2E+2014%2E+The+Human+Communication+Science+Virtual+Laboratory%3A+A+Web+Based+Repository+API%2E+In+The+9th+edition+of+the+Lan)
9. Cassidy, Steve, and Trevor Johnston. 2009. Ingesting the Auslan Corpus into the DADA Annotation Store. In Third Linguistic Annotation Workshop (LAW III). Singapore. [find](https://scholar.google.com/scholar?q=Cassidy%2C+Steve%2C+and+Trevor+Johnston%2E+2009%2E+Ingesting+the+Auslan+Corpus+into+the+DADA+Annotation+Store%2E+In+Third+Linguistic+Annotation+Workshop+%28LAW+III%29%2E+Singapore%2E)
10. Lichtenauer, Jeroen, Michel Valstar, Jie Shen, and Maja Pantic. 2009. Cost-Effective Solution to Synchronized Audio-Visual Capture Using Multiple Sensors. In 2009 Sixth IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS '09). Washington, DC, USA: IEEE Computer Society. [find](https://scholar.google.com/scholar?q=Lichtenauer%2C+Jeroen%2C+Michel+Valstar%2C+Jie+Shen%2C+and+Maja+Pantic%2E+2009%2E+Cost%2DEffective+Solution+to+Synchronized+Audio%2DVisual+Capture+Using+Multiple+Sensors%2E+In+2009+Sixth+IEEE+International+Conference+o)
11. Schiel, Florian, Christoph Draxler, and Jonathan Harrington. 2011. Phonemic Segmentation and Labelling using the MAUS Technique. In Workshop 'New Tools and Methods for Very-Large-Scale Phonetics Research'. University of Pennsylvania, Philadelphia PA. USA. [find](https://scholar.google.com/scholar?q=Schiel%2C+Florian%2C+Christoph+Draxler%2C+and+Jonathan+Harrington%2E+2011%2E+Phonemic+Segmentation+and+Labelling+using+the+MAUS+Technique%2E+In+Workshop+%27New+Tools+and+Methods+for+Very%2DLarge%2DScale+Phonetics+Resea)
12. Sui, Chao, Serajul Haque, Roberto Togneri, and Mohammed Bennamoun. 2012a. A 3D Audio-Visual Corpus for Speech Recognition. In SST2012. Sydney, Australia: ASSTA. [find](https://scholar.google.com/scholar?q=Sui%2C+Chao%2C+Serajul+Haque%2C+Roberto+Togneri%2C+and+Mohammed+Bennamoun%2E+2012a%2E+A+3D+Audio%2DVisual+Corpus+for+Speech+Recognition%2E+In+SST2012%2E+Sydney%2C+Australia%3A+ASSTA%2E)
13. Sui, Chao, Serajul Haque, Roberto Togneri, and Mohammed Bennamoun. 2012b. Discrimination Comparison Between Audio and Visual Features. In Asilomar 2012. Pacific Grove, USA. [find](https://scholar.google.com/scholar?q=Sui%2C+Chao%2C+Serajul+Haque%2C+Roberto+Togneri%2C+and+Mohammed+Bennamoun%2E+2012b%2E+Discrimination+Comparison+Between+Audio+and+Visual+Features%2E+In+Asilomar+2012%2E+Pacific+Grove%2C+USA%2E)
14. Togneri, Roberto, Mohammed Bennamoun, and Chao Sui. 2014. Multimodal Speech Recognition with the AusTalk 3D Audio-Visual Corpus. Tutorial at Interspeech 2014. Singapore. [find](https://scholar.google.com/scholar?q=Togneri%2C+Roberto%2C+Mohammed+Bennamoun%2C+and+Chao+Sui%2E+2014%2E+Multimodal+Speech+Recognition+with+the+AusTalk+3D+Audio%2DVisual+Corpus%2E+Tutorial+at+Interspeech+2014%2E+Singapore%2E)
15. Vonwiller, J., I. Rogers, C. Cleirigh, and W. Lewis. 1995. "Speaker and Material Selection for the Australian National Database of Spoken Language." Journal of Quantitative Linguistics no. 3:177-211. [↗](https://doi.org/10.1080/09296179508590049)
16. Wagner, M., D. Tran, R. Togneri, P. Rose, D. Powers, M. Onslow, D. Loakes, T. Lewis, T. Kuratate, Y. Kinoshita, N. Kemp, S. Ishihara, J. Ingram, J. Hajek, D.B. Grayden, R. Göcke, J. Fletcher, D. Estival, J. Epps, R. Dale, A. Cutler, F. Cox, G. Chetty, S. Cassidy, A. Butcher, D. Burnham, S. Bird, C. Best, M. Bennamoun, J. Arciuli, and E. Ambikairajah. 2010. The Big Australian Speech Corpus (The Big ASC). In 13th Australasian International Conference on Speech Science and Technology, edited by M. Tabain, J. Fletcher, D. Grayden, Hajek J. and A. Butcher. Melbourne: ASSTA. [find](https://scholar.google.com/scholar?q=Wagner%2C+M%2E%2C+D%2E+Tran%2C+R%2E+Togneri%2C+P%2E+Rose%2C+D%2E+Powers%2C+M%2E+Onslow%2C+D%2E+Loakes%2C+T%2E+Lewis%2C+T%2E+Kuratate%2C+Y%2E+Kinoshita%2C+N%2E+Kemp%2C+S%2E+Ishihara%2C+J%2E+Ingram%2C+J%2E+Hajek%2C+D%2EB%2E+Grayden%2C+R%2E+G%C3%B6cke%2C+J%2E+Fletcher%2C+D%2E+Estiv)

*Updated 1 November 2026*