Archive

Archive for the ‘Japanese’ Category

MS Universal Language Input Tool offers correction and transliteration on any web page

Using the UIME, you can “type any language with any keyboard on any web page, using only the Roman characters present on every keyboard.”

ms-universal-Language-Input-Tool

And you can install your favorite input language in your web browser, like so:

ms-uime-add-japanese-ie8 ms-uime-add-japanese-ie8-test

Sample of how students use the foreign language TTS (text-to-speech) in the LRC

Last week, I noticed this student in the LRC working on a speaking assignment in her SLA class. First, she wrote a draft of her presentation. She then had the Deskbot TTS wizard (in the lower right of her screen) read out the draft to her. She recorded her version, modeling after the Deskbot’s pronunciation. Then she had the Deskbot read out the draft again, to compare with her own recording.

Look for the little guy in the lower right corner of the screen with the cartoon bubble over his head.

I favor the use of the Deskbot TTS (a Windows XP technology) as an easily accessible speaking dictionary, including during face-to-face teaching, when students may otherwise be too shy to make an utterance because they are not sure about the pronunciation of a single word, or even ask the teacher to pronounce it for them. Prosodically, the deskbot TTS leaves many things to be desired. Let’s hope that Windows 7 will enable us to set up more advanced TTS support in the LRC.

Learn Chinese character stroke-order with slowed-down animated GIFs

Further to our prior tips on learning Chinese stroke order, now you can take your time, in the LRC: To facilitate your practicing of Chinese character stroke-order, we have used the most helpful site (also available by direct download) created by Tim Xie for the California State University, Long Beach, to create 100 different speed versions, and one comic strip like static image, for each of the several hundred of animated GIFs demonstrating Chinese character writing, and made them available on the LRC computers under Internet Explorer Favorites – Example:

a1f5_strip

You can access the files with the stroke order speed of your preference from the LRCCOED434 student computers, like so:

chinese-stroke-order-animated-html

(Many thanks also to the authors of programmable ImageMagick image editor and corresponding Unix shell scripts that we could use in the production of the slowed down animated GIFs. To create your own version of these slowed down animated GIFs, or others similar websites, feel free to pick up and/or adapt our shell script here).

Speaking/Listening Assessments and Oral Exams: A comparison what the LRC has to offer

  1. Moodle:
    1. I proposed for installing one of the free audio recorder plug-ins into our Moodle, but we are not there yet.
    2. However, we do have a new video recording assignment (which is based on Kaltura).
      1. Format: free form, according to your written instructions in the assignment. Students can review and repeat the recording as often as they wish.
      2. The video overhead is minimal since it is streamed, and video is better for authentic language assessments – unless you specifically prepare your students for phone interviews: then just have students step out of the viewing angle of the webcam).
      3. thanks to Moodle, the familiar interface and the underlying LMS support infrastructure, it is easy
        1. for the teacher
          1. to create and assign a video-assignment
          2. to grade it from the gradebook
        2. for the students to take it and submit it.
      4. LRC Support:
      1. Since our PCs have no built-in or added webcam (proposed), we can currently only use our 5 iMacs for Moodle video assignments. Since the MACs do not have headsets (but built-in microphones), the audio quality is not as good as on the PCs. Since 5 seats are not sufficient for class-size activities/exams, it is best to use this as a homework assignment  
      2. I list all necessary steps for a video assignment
        1. here for teachers
        2. here for students.
        3. all my Kaltura posts
      3. Additional support is available through the campus Moodle support team.
  2. Sanako
    1. Sources
      1. Dual-track comparative recorder:

        1. the teacher can prepare an input track (or provide one live. Preparing is easy, and worth your while, since easily reusable. I can help you)).
        2. the students records on her own track
      2. Pair  and group recording:

        1. Sanako makes it easy to pair or form groups of students and to record free-form conversations.
        2. These recordings can be either controlled remotely by the teacher or locally by the user
    2. Control
      1. Remote-controlled recording under exam conditions,
        1. responding to a listening cue within a preset (or live) pause in the teacher track
          1. model&imitation:  for phonetics and pronunciation exercises,
          2. question&response for a wide variety of activities as commonly used in SLA textbooks and classroom, including practicing grammar structures or vocabulary recently
          3. question&response&model response: the teacher can also include after the pause in the teacher track a model answer for the students to compare their own output to.
        2. automatically saved with student names to be accessed from the teacher office desktop
        3. easy comparative grading using Audacity (see below)
    3. User-controlled recording is also possible, using the student recorder in manual operation mode
      1. which has more language learner features (bookmarks, voice graph, dual band recording), and  a simpler  interface than a full blown audio editor like Audacity (see below).
      2. The task how to save and sent the assignment to the teacher is here left to the user.
    4. LRC support:
    1. I can help you
      1. creating an audio recording with your content and speaking cues and pauses – using Audacity (see below)
      2. conducting the remote-controlled exam
    2. Up to 20 seats can take an oral exam simultaneously, until we get more Sanako licenses. However, we  found a way to split classes into 2 halves and have consecutive exams (we can play audio on the other students’ headsets to provide for exam conditions). The LRC main classroom is equipped with 30 seats for 2 consecutive exams with Sanako headsets.
  3. Voicethread is a popular online recorder, especially  for educational institutions that have no onsite support.
    1. Visual and audio cues can be provided by the teacher.
    2. Pairing of students has been attempted via sharing and responding/commenting to the partner’s submission. This is not a realistic conversation.
    3. Recordings are stored in the cloud.
    4. There is no integration with the SIS (accounts – getting students set up with accounts that can communicate back with the teacher is a challenge) and LMS (the Moodle integration is superficial). 
    5. Voicethread is not free. The ELTI, however, has a subscription. LCS does not.
    6. LRC Support:
      1. We support Voicethread exercises with new and improved headsets.
      2. Help is available through the vendor.
  4. Audacity:
    1. for teachers and LRC staff and other language professionals:
      1. best free audio editor, also good for comparative grading. I routinely make my audio exam recordings with Audacity.
      2. LRC support: I have tips and tricks how you can use it in your teaching preparation and grading.
    2. for language learners: not the recommended option, since Audacity has not a feature set geared towards language learning nor support for language assessment workflows:  
      1. Language learners do no need an audio editor for speaking exercises, they need a recorder. If you are a language learner, it is not pedagogical to be able to technically edit and refine your audio recording. Rather rehearse, reflect on and repeat your audio recording, until you are happy with your language output.
      2. Audacity is too technical: It involves too many steps, options and settings for the students to record, save, export and name the audio and to get it to the teacher, and (if it is not uploaded into a Moodle assignment, which could then be a Kaltura  assignment anyway, see above), too tedious for the teacher to manage and grade files.
      3. LRC Support:

        1. If your students are technically inclined, we do have Audacity installed in the LRC.
        2. Your students should not find it difficult to read the documentation. Here are my posts on Audacity.

How a Student takes a Moodle Video Assignment in the LRC

2011/09/22 2 comments
  1. On one of the LRC iMacs, in the Safari web browser (open new window with COMMAND-key+n),
  2. go to your Moodle course, (1) find the video Assignment, read the assignment instructions (what your teacher wants  you to record). Then click underneath the  (2) button: “Add video Assignment”, to open the (3) submission window:  moodle-add-video-submission
  3. Note that she will also have to allow the flash player to interact with her webcam first.
  4. kaltura_thumb1 If you see no web cam video window, only a black frame, read in.
  5. In the submission window, choose the tab “Webcam” (1), use the dropdown to select the camera hardware (2). moodle-kaltura-webcam-tab-camera-dropdown
  6. Check the headset microphone audio: The external headset microphone on the iMacs did not work., but now it does, provided you do this: control-click on Flash’s a video preview window (= the window where you see yourself like in a mirror) for the  web camera, and click on “settings”.
    •  
    • Click on the microphone icon :
    • CIMG0021
    • Make sure the USB PnP  device is selected.
    • CIMG0022
    • You can bring up the settings dialogue, make sure the USB PnP device is chosen for audio and CRANK up the microphone input sensitivity! Then, by tabbing (don’t speak yet) on the headset microphone, test the volume levels with the built-in volume meter (should  show lots of green bars when you tab). Unlike in the picture, do not choose “reduce echo ”.

    student-recording-CIMG0015_thumb1

  7. Start (3) the video recording.
  8. Afterwards, the student can review (4) her submission.
  9. If you don’t like your first recording, (3) “record” over it and review again with (4) “Play”. If you do this and the video appears frozen, drag the timeline cursor forward to get the re-recorded video to play. If this does not seem to work, you are likely still able to submit your 2nd attempt, just not review it again.  kaltura-timeline-drag-me1
  10. Click through all the “Next”etc. buttons:
    1.  pauline-moodle-kaltura-next3 pauline-moodle-kaltura-next4CIMG0070CIMG0071
    2. until you get to the feedback page for the student (caveat: in IE9, the video does not fit into the frame provided on the page). student-assignment-result
    3. Note, it warns you to be patient now: kaltura-teacher-upload-student-upload-combineda
  11. LRC support:
    1. Depending on your hardware (webcam), software and network support, you can record your language speaking video assignments on any device that has a webcam and a browser that supports flash – and even more devices, if you are willing to post process and upload the video clip.
    2. If you run into problems or want to use a tested setup, we recommend using the LRC. Since our PCs have no built-in or added webcam (proposed), we can currently only use our 5 8 iMacs (see LRC Layout, see Classroom Calendar and iMacs Calendar for availability). Our (limited) tests worked better in Safari than Firefox.

Successful test of the new streaming video recording assignment for language proficiency assessment in Moodle.

2011/07/20 1 comment
  1. Video recordings allow for a more authentic assessment of spoken language proficiency. Today, we could test a new Moodle video assignment type for Almut’s Summer II Elementary German II class: Preparing a natural sounding statement on one’s favorite holiday. This Moodle assignment type is an extension to students of the teacher video upload tool using a service Kaltura – we hinted at that earlier.
  2. Teachers, when editing their Moodle course,  can find this assignment type in the dropdown: Activity as “Video”. dropwdown-activity
  3. The assignment options: Note that creating a video assignment is pretty much the same as creating other assignments in Moodle, and so is grading, except instead of reading, you view the submission, right within the web page.  kaltura-assignment-options
  4. An example of guiding questions for a recording assignment is here (topic: holiday, language: German): moodle-video-assignment-guiding-questions
  5. Here is how the teacher sees the student submission in the gradebook.
  6. kaltura (1)
  7. teacher-grading-roster-CIMG0021
  8. Caveat: in Safari, we were not able to close the video popup, after viewing it from the class roster, and could only back out of the entire gradebook. Instead, from the roster, first open the page with the individual student submission, and review the video there on that page:
  9. teacher-grading-CIMG0027
  10. For the student experience, see here.
  11. Further reading: The CTL has a number of step-by-step instructions which we recommend for further reading: Student Video Assignment, and specifically for students: Upload a Video for a Video Assignment; for instructors: Grading Video Assignment Submissions.
  12. The LRC had originally prepared to record the students with our old, handheld mini-DV cameras, import the movie into iMovie or Windows Movie Maker, then find a way to get the files (with easily identifiable submitter names) to the teachers. Hitherto, our best option was compression of the video to to fit into the Moodle 64MB file upload size limit  (which, even if you decide to shoot and produce your video elsewhere and bring it as an uploadable  file to the assignment , does not apply to the Video assignment either).
  13. Preliminary testing seems to indicate that video recording of pairs/dialogues is also possible with the LRC’s webcam setup . However, because of the angle restrictions, capturing such sessions will be less natural.
  14. Overall, the new Moodle video assignment seems a major improvement for  all parties – students, teachers and support –, and can help with more authentic assessment of

Speaking dictionary/pronunciation help, Part 1–the pedagogy

2011/05/24 1 comment

On the LRC PC’s,  you can copy foreign language text to the clipboard (or with non-western languages to an MS-Word file) and have a wizard (right click, choose: “Read Clipboard”) pronounce for you, like so:

student_writing_mey_halffilled_selected_deskbot_merlin_cropped

The following languages are supported:

Language Demo
American English  
French Le Monde
German  
Italian  
Japanese Japanese
Korean  
Portuguese (Brazilian)  
Russian Gazetta
Spanish El Pais@10:34

Note 1: In reality, the audio quality is smoother than in these demos which needed to be recorded over the network using MSTSC, because of Windows Media Encoder not being able to capture the special video overlay of the MS-Agent technology.

Note 2: Deskbot has problems understanding the Windows clipboard text encoding of some non-western languages. For these languages, instead of simple double-clicking the deskbot wizard to have him read out the clipboard (which would result in gibberish), paste your text into a MS-Word document and have deskbot read this document for you instead: Use the Deskbot option “Read Word document” (allowing Word to translate the character  encoding into a format Deskbot can understand enough to aurally (even if not visually) display).

Listen to the Deskbot’s self-introduction when you log in to learn how to change the default language to the language you study.

You can also change other settings (in the LRC automatically limited to your current user session) by right-clicking the  deskbot notification-area icon deskbot-notification-area-icon

Or if you want to configure your own Windows XP computer to support foreign language text-to-speech, the procedure and settings we recommend are documented in Part 2.

Speaking dictionary/pronunciation help, Part 2 – The technology: Installation & configuration of the free Windows XP text-to-speech wizard

  1. MS-Speech synthesis which comes free with Windows XP is a mature technology which gets even better if combined with free downloadable voice/language combinations and a simple interface.
  2. Since 2005, I have set up the following text-to-speech environment in language learning centers, wherever I could: The pleasant, speedy interface and convenient integration into the computing productivity environment (Clipboard, Internet Explorer) seems to lead to higher usage than online speaking dictionaries. And the price is hard to beat (all components are free on Windows XP).
  3. The Deskbot pronunciation – most of my own testing has been for German and English, but  I rely on the feedback of native speaking other teachers and tutors for other supported languages – is good on a word level . The prosody of sentence pronunciation leaves a lot to be desired, but is a start nevertheless.
  4. If you want to install the Deskbot Text-to-speech wizard for your users, or for your own testing and language learning, here are the steps and settings we recommend:
  5. Use these installers:
  6. Language TTS Engine Language Add-On
    American English tv_enua.exe already installed
    British English lhttseng.exe already installed
    Dutch lhttsdun.exe agtx0413.exe
    French lhttsfrf.exe agtx040C.exe
    German lhttsged.exe agtx0407.exe
    Italian lhttsiti.exe agtx0410.exe
    Japanese lhttsjpj.exe agtx0411.exe
    Korean lhttskok.exe agtx0412.exe
    Portuguese (Brazilian) lhttsptb.exe agtx0416.exe
    Russian lhttsrur.exe agtx0419.exe
    Spanish lhttsspe.exe agtx0C0A.exe
  7. Move the Deskbot near the lower right corner notification bar to have it not clutter the user interface.
  8. Do not automatically read clipboard when changed
  9. Do NOT play animation sound effect: effects wear out quickly and become distracting or annoying.
  10. Default set to American male: or whatever in your shared environment  is the language all learners understand .
  11. You have to click “set” to actually effect a change, and close the options dialogue.
  12. Set read random greeting at startup: in a shared computer setting, you can on startup.
  13. Have the Deskbot introduce himself with basic usage instructions: In order to have this feature self-advertise in the LRC, beyond my face-to-face-classes in the language center, I have the Deskbot read out an informational welcome message every time it starts up (you could use this feature also to have the Deskbot issue a random greeting in your target language).  
  14. Show balloon – useful since words scroll in as spoken.
  15. In the Speech control panel, you can set the speed to slow – useful for language learners.
  16. MS Agent Options: TBA
  17. You can view the Deskbot install process in sequence as a screencast:
  18. Or you can view the configuration options screens with recommended settings in this screenshot gallery: 
  1. For my own – and maybe your – reference, here is a raw screencast of the most recent installation and configuration of Deskbot multilingual text-to-speech engine for MS-Windows.
  2. time

    content

    0,00

    Base install begin

    2,50

    Deskbot – including its installer – cannot handle the character encoding of some non-western languages that it speaks (Japanese, Korean, Russian). In order to have Deskbot speak these languages, you have to use the Deskbot option “Read Word document” (allowing Word to translate the character  encoding into a format Deskbot can understand enough to aurally (even if not visually) display).

    4,13

    fine tuning for language learning purposes in deskbot options and speech control panel after install

    6,09

    “advanced character options”: set the speaking speed

     

    Returning to the base installation of some more languages

    8,04

    more deskbot options

    10,34

    test of Spanish pronunciation with elpais.com

  3. In a computer lab environment with Faronics Deepfreeze and Symantecs Ghost imaging, you will also want to copy these settings to the Default user, and make sure that the deskbot startup items gets added to all users or HKLM registry hive (the interface does not allow for installing for all users).