Speech Recognition with Mobile Devices

Scenario – Speech Recognition with Mobile Devices

The potential for speech recognition on mobile devices is significant and already practical with new apps such as Dragon Dictation and Dragon Search despite limitations with the current free version. Human speech is processed and recognized through online servers that provide opportunities for inputting text without having to type. Learners who have difficulty typing (or cannot type at all) with the on-screen keyboard can quickly convert their speech into text for applications such as e-mail, taking notes, creating blogs, and searching information on the Internet. Currently the Dictation app is limited to 30 seconds, although it is possible to append the text for each 30 seconds until you have completed your dictation.

Software: Dragon Dictation and Dragon Search

Hardware: iPod Touch with external microphone; Wi-Fi access required (or iPhone)

Process

• Attach an external microphone to the iPod touch.
• Open the Dragon Dictation App
• Press Tap and Dictate and speak slowly and distinctly with minimal background voice.
• What you say up to 30 seconds will be processed and displayed on your iPod screen.
• You can either edit, copy, or e-mail the text.
• For Dragon Search say your search terms, which will be converted into text and then submitted simultaneously to YouTube, iTunes, Wikipedia, Twitter, and Google.

Dragon Dictation - Converts what you speak into surprisingly accurate text. You will need a microphone for your iPod Touch of course. The process is simple: press Tap and dictate and start speaking - slowly and distinctly with minimal background noise works best. After 30 seconds Dragon Dictation will process your speech through online servers and display the converted text on your iPod screen, which you can either edit, copy, or e-mail. Although a bit cumbersome you can speak in segments of 30 seconds, appending your text each time so you can dictate for longer periods of time. Although the app is free for now, there is bound to be a paid version with more features.

Dragon Search - Processes search results based spoken text with impressive accuracy, although certainly not perfect. As with Dragon Dictate you will need a microphone for your iPod. The process is simple: press Tap and Speak and say your search phrase distinctly with minimal background noise for the best results. Press the Done button and your phrase will be searched by Google, Wikipedia, YouTube, iTunes, and Twitter. For example, I spoke "Marcellus Shale" and Dragon Search found my tweet sent only minutes before - amazing.

Instant Answers

You can edit spoken text in the search box if Dragon Search does not interpret the search phrase as you expect. For instant answers dictate terms according to Google Search features http://www.google.com/intl/en/help/features.html . Try this example: say "How many inches in a mile", "72 inches in centimeters" or "2 tablespoons in ounces" to trigger Google Calculator, which immediately calculates and displays the results as shown below. Google Calculator does not recognize terms such as "square root of 987", although other web sites such as WikiAnswers will provide the answer once you open the page.

Reading ebooks

Scenario – Reading E-books

Electronic readers provide students access to an enormous library of free and electronic books, periodicals, and newspapers that can be directly read on an iPod. Although the use of e-books is already growing, there is expected to be a substantial impact when the iPad and other touch screen tablets reach the market in early spring. All students benefit from reading electronic materials on mobile devices and avoiding having to lug around heavy books. iPod e-readers such as Stanza provide a number of benefits for students with assistive needs. Although the screen is small, students with visual disabilities can change the font, color, background, margins and brightness of the screen to improve visibility. In addition students prefer to read books on an iPod rather than drawing attention to a special-purpose device.

Software: Stanza Desktop for Windows or Mac; Stanza App for iPhone

Hardware: iPod Touch; Wi-Fi access required

Process

• Download the free Stanza iPod app.
• Run the Stanza iPhone app and use Catalog to Get Books or Downloads to download books in ePub/eReader format from a URL (e.g. http://www.epubbooks.com/book/312/hugo-hunchback-of-notre-dame.epub)
• Download the desktop app from http://www.lexcycle.com/stanza and open a PDF, text, HTML, or Word document. Choose Tools > Enable Sharing and choose Shared then Books on your computer to download the open document.
• Choose Catalog to open your downloaded and converted e-books ready for online reading! In Stanza there options to annotate text, look up word definitions, and customize the fonts, color, layout, and brightness of the text.

Mobile Text to Speech

Scenario – Mobile Text-to-Speech

Although desktop software such as Flame Reader and iSpeak It! provide more features, control, and organization of text-to-speech recordings, they typically require a three-step process: convert text to speech, copy the recording to iTunes, and copy the spoken text to an iPod touch. Apps that directly convert text to speech as it appears on the iPod screen can provide advantages over desktop conversion and copying depending on how they are used. In addition to enhancing comprehension and listening at the same time, mobile text-to-speech also enables people who are unable to speak to communicate in person or by phone (iPod users can play the audio over a telephone handset).

Software: iPod touch Apps such as Speak It, Web Talk, and Read 2 Me

Hardware: iPod Touch

Resources: http://teachlearntechblog.blogspot.com - under Menu click Assistive Mobile Technology to Help Students Succeed.

Process

• Run the app Web Talk, which has a built in browser and open the web page you want to read.
• In preferences adjust the voice, volume, and speed controls.
• Press the Play button to listen to page's spoken text.
• Use the app Speak-It! to read typed or copied text in the text to say window. Two American and two British voices. Spoken text can be saved but not exported.
• Read 2 Me allows importing of text files, although they must be in URL format.

Description of Speak-It!

Type or copy and paste text in the "text to say" window and have a synthesized voice speak back to you. You can choose from two American voices (good quality) and two British voices (better quality), increase or lower the volume, and adjust the playback speed. There are a number of application for Speak It! including hands-free listening of online news, playing back text for those who are unable to speak, and assisting auditory learning with reading text. The app currently cannot upload files nor export spoken text as mp3 files. Speak It! lacks the ability to remember current playback position and does not allow scrolling of recorded spoken text. For long text such as chapters and books, I recommend a solution such as iSpeak It and iTunes.

Listening to Blogs

Scenario - Listening to Blogs

Blogs provide opportunities for students to read, reflect, and write in an open environment for learning through feedback, revision, and sharing. Converting them to spoken text enhances learning opportunities in different ways by empowering students to:

• Listen to blogs if they have a visual impairment.
• Read and listen to blogs at the same time to help improve their reading comprehension.
• Listen to their writing as a tool to spot repetitions, choppy phrases, awkward sentences, and misspelled words. Listening to spoken text can help students detect lack of rhythm and sentence variation in their writing.

Students can subscribe to the spoken text in each blog post as podcasts, which can synced to an iPod touch for mobile listening, although the text is not copied.

Click the Listen Now button at the top of this post to evaluate the quality of spoken text.

Software: Blogger, WordPress, or TypePad; http://odiogo.com

Hardware: iPod Touch (or iPhone)

Process for Blogger (works similarly with WordPress and TypePad)

• Go to http://odiogo.com and submit your Blog URL and e-mail address.
• Click the Add Odiogo Listen Button Widget
• Log on to your Blogger account
• Select the blog that you want to convert to audio.
• Click the Add Widget button.
• Rearrange the Subscribe to My Podcast Gadget in the Layout menu.
• Click the Save button.
• Click the Listen Now to hear your blog!
• Click the Subscribe to My Podcasts button.
• Click to Subscribe to the blog's podcast in iTunes. (audio-only)
• Copy the blog Podcast to an iPod touch and listen to the posts!

Recording Class Lectures

Scenario – Recording Class Lectures

A number of instructors are recording their lectures for course casting that provides opportunities for students to review materials for reinforcement or that they missed. If course podcasts are not available, then students can request permission from their instructor, especially if they are registered through their College's office of Accessibility Services. Equipped with an inexpensive microphone an iPod touch can be used for recording lectures, which can then be transferred to iTunes for conversion and tagging. Lectures can be organized by playlist, topic, instructors, and date to help students find materials for repeat listening and review. The recordings in turn can be copied to an iPod Touch to allow mobile playback using the same method of organization.

Software: Apple's Voice Memos app, iTunes for Macintosh or Windows http://www.apple.com/itunes

Hardware: iPod Touch Gen 2 or Gen 3 (has volume slider bar on side) for recording; external microphone

Process

• Attach an external microphone to the iPod touch audio port.
• Open the Voice Memos app and start the recording.
• Stop the recording, trim start and end as necessary.
• Connect iPod to your computer and copy recordings to iTunes.
• Tag the recordings.
• Copy the recorded chapters to the iPod Touch. Students can read and listen at the same time or listen only.

View PDF Tutorial (opens in new window)

Voice Memos – Apple provided this app as part of iPod Software Update version 3.0, which requires a 2nd or 3rd Generation iPod and a microphone with a 3-ring 3.5mm connector. Voice Memos allows recording, trimming, saving as a voice memo, and sharing as an e-mail MPEG-4 (.m4a) audio attachment.

Feminist Agency

The Women’s Studies Program Presents: The 16th Annual Julia Reinstein Symposium "Celebrating International Women"

Keynote Address by Dr. Jessie Kabwila Kapasula

“Feminist Agency in the Age of Beyonce's 'Single Ladies':Faces of International Feminism in Transnational Popular Songs”

The keynote address refers to three music videos, which can be played by clicking the links below. The referral times in the address are in parentheses.

Keynote address

Music Videos

Beyonce All the single ladies (16 min. 12 sec.)

Music Video: http://www.youtube.com/watch?v=4m1EFMoRFvY

Lyrics: http://www.metrolyrics.com/single-ladies-lyrics-beyonce-knowles.html

Neyo Miss Independent (38 min. 51 sec.)

Music Video: http://www.youtube.com/watch?v=k6M5C-oKw9k

Lyrics -http://www.metrolyrics.com/miss-independent-lyrics-neyo.html

Nalu House Money and Car (42 min. 1 sec.)

Music video: http://www.youtube.com/watch?v=moex19whag8

Converting PDF's to Audiobooks

Scenario - PDF's to Audiobooks

For many students comprehension of printed text materials depends on reading and listening to the text at the same time. Although it is possible to scan printed materials into electronic format the process is tedious and inaccurate. Some publishers will provide audio materials in a format that requires a special player such as a Victor Reader. Typically these players are cumbersome to use, bulky, and are not integrated within a student's learning environment. There is also the stigma of having to carry around the device. In this scenario publishers provide searchable PDF files usually by chapter.

Attention! Conversion of copyrighted materials can only be used by students who have purchased the textbook and are registered with the office of accessibility services. The PDF’s and converted spoken text cannot be distributed.

Software: PDF to text converter; iSpeak It! for Macintosh or Flame Reader (Windows); iTunes for Macintosh or Windows http://www.apple.com/itunes

Hardware: iPod Touch for reading and listening; portable media for audio only.

Process

• Open the PDF document - typically a chapter file - and convert it to text (Macintosh Preview, Adobe Acrobat, CutePDF for Windows, or http://zamzar.com).
• Copy and paste the text into FlameReader for Windows or iSpeak It! for Macintosh
• Convert text into MP3 or AAC audio format.
• Add the recording to iTunes and then tag the audio recording.
• iSpeak It for Macintosh provides options to facilitate tagging, bookmarking, and categorizing the spoken text as an iTunes audiobook. In addition the text is copied to the Lyrics tab.
• Copy the recorded chapters to the iPod Touch. Students can read and listen at the same time or listen only.

Converting Text-to-Speech for iTunes and Mobile Playback

View PDF document (opens in new window)