Saturday, February 1, 2020

Some Tricks to Ease Language Sample Transcription Pain...

I do a lot of language sampling and analysis as part of evaluations- whether a more straightforward speech/language or social cognitive evaluation, it's an essential part of showing current functional strengths and weaknesses. And you know how much I dig narrative.

Periodically, I try to do some searching to see if there is anything tech-wise that can ease what can be some pain within this process. Speech-to-Text (STT) technology, in which a device or some kind of software is used to transcribe spoken language, continues to evolve. Best I know, we still can't play a recording into STT tools (such as those built into a Mac or Google Docs) and have it accurately transcribe. However, I found this idea recently and gave it a try, with nice results. Paraphrasing Leah Fessler of Quartz at Work (thanks for this post!), the steps are:

1. Find a quiet space.
2. Plug yourself in (you may benefit from using headphones with a mic).
3. Open a blank Google Doc (Note: only in Google Chrome)
4. Open the Voice Typing tool (Tools>Voice Typing)
5. Ensure the Voice Typing button appears.
6. Ensure your microphone is turned on and your language is set.
7. Listen to a small portion of your language sample recording (e.g. a sentence)
8. Click the microphone button, and repeat what your client/student said. Remember that you can say "comma" for commas and similar for other punctuation.
9. Turn Voice Typing off as you listen to the next part of your recording (whatever you can hold in your short term memory), then continue, repeating steps 7-9.
10. Watch along as you transcribe and make corrections and additions.



A few tips may be of help:
-I like the Voice Memos app for iPad. If using the app of the same name (this one native to the operating system, so free) on iPad or iPhone, make sure you turn your device Auto-Lock (in Settings>Display and Brightness) to something longer, like 5 min. I notice after auto-locking, my phone's Voice Memos app moves the recording back 20 seconds or so, which is annoying.
-You can use either app's 15 sec forward/back buttons liberally to re-listen or go forward.
-Finally, I think the above steps saved me time and boredom in the transcription process, but I used step 10 extensively. If you have a student with a lot of revisions or repetitions, you'll find you are better off doing a combination of typing and dictating (using dictation with the student's more fluent sentences).

Let me know if you try this and what you think, or if you have other strategies!

4 comments:

  1. This is something very important and useful topic to discuss. Great article.

    ReplyDelete
  2. Wish you all the best for your new upcoming articles and kindly write on topic Ecommerce website and also on Design Logo
    thanks.

    ReplyDelete
  3. This comment has been removed by the author.

    ReplyDelete
  4. More and more people are learning how voice to text and speech recording can save them time and work. It quickly types out spoken words, so you can use it to take notes, write content, and do other things at the same time. There are now more languages and accents that AI can transcribe. This easy and open-source technology makes it easier for people with disabilities or who would rather talk than type to get their point across.

    ReplyDelete

 
.