speech-recognition | Thomas' Work Space

“"Mira, mamá! Sin manos!". Can speech recognition tools be soundly applied for L2 speaking practice?” FLEAT VI, August 11-15 at Harvard University

2015/08/13 plagwitz Leave a comment

2015/04/06 plagwitz Leave a comment

Since I have been asked to advertise this:
1. These Speech recognition tools, while an invaluable partial evaluation automation or AI tool, come free with MS-Windows (7 Enterprise/Ultimate, 8 all distributions) – crucial for often resource-strapped language learning environments.
2. The tools do not come with pre-built exercises – a blessing in disguise since this makes them easier to integrate them with your existing textbook-based syllabus. I provide some ideas how you can easily derive speech recognition exercises from your syllabus/textbooks.
I recommend having a look at the entire new issue of FLTMag, the magazine on technology integration in the world language classroom, which
1. examines other software that supports second language learning with video conferencing, support for writing and vocabulary learning (on tablets versus print materials) ,
2. reviews a book on “Ideas for integrating technology in the classroom”
3. and advertises upcoming conferences. Hope to see you there.

Categories: research Tags: speech-recognition

2014/10/26 plagwitz Leave a comment

For your convenience, I provide two versions of this presentation: the slide deck:

And a screencast of my presentation:

(Unfortunately, PowerPoint 2013 seems to have introduced a bug in alignment of narration and video. If they fix it, I will republish),

2014/09/14 plagwitz Leave a comment

My initial test of W7ASR mV8MRY used an end-of-chapter web quest from a 1st-year German textbook. From reading for the gist to pronouncing an authentic target culture text is a challenge – make it extra credit.
Turn existing writing exercises into speaking/ reading/ corrective writing exercises: Free form writing during note taking, e.g. when answering comprehension or short essay questions, fits the continuous speech optimization of W7ASR.
You will likely see reduced recognition accuracy and less proofing feedback, but gain wider applicability for (grammar, but especially vocabulary) drills that use discrete writing, e.g. filling in cloze exercises (navigation need not be done by voice), even in a web browser from your LMS or online textbook.
Turn “flipped classroom” homework preparation fill-in-the-blank conversation suggestions into more hands-on phrase dictations before your in-class communicative practice.
Similarly, my Spanish screencasts, e.g. qK9fKV, flips our Spanish for Law Enforcement: If W7ASR can ”understand” your Miranda warning, a suspect likely can also.
Finally, the most advanced language learners can replace writing of essay tasks from their textbooks by speaking them – use of only ASR for editing is optional.
All above tasks are based on existing textbook and/or syllabus which are easily extended by Windows 7 “open response” speech recognition.

2014/09/14 plagwitz Enter your password to view comments.

2014/09/13 plagwitz Leave a comment

Tools needed for my LRC speaking practice task design using Windows7 Automated Speech recognition:
1. proofing tools and tracking changes (suggest MS-Word),
2. a screencast recorder (free MS-Community Clips: MGFGX3),
3. an LMS for uploads of one to several dozen MBs files),
4. and the W7ASR Profile Tool (to back up individual voice training data in a shared, "frozen" LRC).

2014/09/12 plagwitz Enter your password to view comments.

2014/09/07 plagwitz Enter your password to view comments.