Task 3 – Voice to Text Task

Here is my transcribed dictated assignment.  I didn’t tell a story per se, but rather shared a few thoughts I had about the topic this week.  I apologize for the rambling nature of the dictation, but even though I had a fairly organized outline of what I wanted to say, I still found it difficult to talk smoothly and coherently without a script.

I used MS Word on my Android phone with the built-in transcription feature, and then converted it into PDF.

Etec Task 3 transcription

 

Here is the analysis:

  • How does the text deviate from conventions of written English?

Proper written English has quite prescriptive conventions and structure to follow in terms of grammar, sentence/paragraph structure, punctuation, capitalization, etc.  I use the voice to text feature quite often on my phone, and have become fairly proficient with inserting simple punctuation such as commas, periods, and question marks.  However, there should be quotation marks in some places, and I was unable to organize my essay into paragraphs – I’m sure there is a way to do it, but I wasn’t able to figure it out for the purpose of this assignment.  A properly written essay would be organized into structured paragraphs each with their own individual topic.  My transcription also has missing capitalizations (such as for the names of the text chapters we’re reading), run-on sentences, extra inserted words, and incorrect tense usage.  Written English usually has a coherent flow of thoughts that progress in sequence because one has much more time and opportunity to organize their ideas whereas this text is much more disjoined and random, just as spoken conversation might be.

  • What is “wrong” in the text? What is “right”?

I am quite surprised that the majority of what I said was actually transcribed quite correctly – at least for the words spoken.  I did highlight some of the major errors – the first one being my name.  There are some missing capitals – the names of our text books should be capitalized.  That being said, the program does recognize proper nouns such as Russian or English.  Some words were incorrectly understood such as ‘fanatic’ which should have been ‘phonetic’.  There were extra words inserted such as ‘oh’, probably inserted when I paused and mumbled to collect thoughts.  Overall, the majority of the words were transcribed correctly, but the text seems quite rambling, disorganized, and chaotic.  Grammatically it may be mostly ‘correct’, but it doesn’t flow smoothly when one reads it.

  • What are the most common “mistakes” in the text and why do you consider them “mistakes”?

Probably the most common ‘mistake’ I can see is the insertion of random words such as ‘oh’ although it didn’t happen too often.  If the program is truly transcribing my speech, then it probably isn’t a ‘mistake’ per se because when I speak I tend to insert thinking pauses such as ‘umm’ or ‘oh’, so it may have picked that up.  There are also a couple of tense mistakes – used instead of use, and pass down instead of passed down.  These are mistakes because they are grammatically incorrect according to conventional English language.  Mistakes in transcribed text probably occur mostly when incorrect words are inserted rather than the originally spoken words because of the phonetic differences in certain syllables, and the similarities in certain words (for example ‘by’ and ‘buy’).  I have noticed that often when I dictate, the program will self-correct based on context, so accuracy of these programs is continually improving.

  • What if you had “scripted” the story? What difference might that have made?

If I had been able to ‘script’ my story, it would have made things much easier because I would have had the opportunity to organize my thoughts ahead of time into a much more coherent, organized flow of ideas.  There may still have been mistakes in terms of incorrect words, punctuation, or grammar, but it would probably be much easier and smoother to read.  That being said, it would lose the ‘conversational’ element of this assignment.

  • In what ways does oral storytelling differ from written storytelling? 

Oral storytelling has a much more personal and animated element to it because the person telling the story can add vocal inflections and gestures which can add a lot to the ambience and spirit of the story, engaging and involving the audience much more actively.  Oral storytellers can use physical props to aid their tale. Each time the story is told, it might change a bit depending upon who is telling the tale, or what elements are emphasized.

Written storytelling depends upon the ability of the author to use the appropriate descriptive words in the correct sequence to paint a mental picture for the reader.  It is not as personally interactive as oral storytelling, but can be equally engaging if well written.  Instead of hearing the voice of the storyteller, the audience inserts their own voice.  Physical props are not possible, but pictures may be inserted to help establish the scene.  In the case of a written story, there is no change.  Each time it is read it is the same and there is no variation except in the mind of the person reading it.

4 Thoughts.

  1. Hi Marlis,

    It’s interesting how you were already used to speaking with punctuation, obviously already used to the tool. I used Google as well for my dictation, but since I don’t use the tool often, didn’t add in periods, commas etc. It’s so often HOW you use a tool! I did find it interesting that my dictation automatically took out the “oh’s” and “um’s”, but yours didn’t. I do wonder what and how it’s trained to take those sounds out. What else is it automatically deleting? I completely agree with your comparison of oral storytelling to written text. The amount that you get from the teller themselves is so powerful. I wonder if that’s why illustrations in text can be so powerful.

    • Hi Christina, great thoughts. I think there’s actually a lot more that can be done with the dictation feature if one has the time to dig into it – it’s actually quite powerful and useful. I’ve only gotten to the basic punctuation (commas, periods, question marks, etc), but I think you can actually command it to do ‘returns’ (start a new paragraph), parentheses, etc. I just haven’t figured out the proper commands yet. As I said with Emily, I’m not sure how it differentiates between the punctuation command and the ‘word’ – I think that still has to be manually overwritten.
      Since I use the tool quite frequently, I consciously try to avoid the ‘umms’ and pauses that come with normal speech, and I was speaking quite slowly and deliberately, but it still caught those errors. It might have just been my particular pattern of speech. It does have quite a powerful auto-correct feature, and seems to pick up the context and sentence structure of your intent quite well. For instance, one might say ‘The dog ran out in the rain’, which might initially translate to ‘The dog pan in pain’, but the algorithm picks up that this doesn’t make sense and corrects it. A powerful tool, and definitely a game changer for recording text.
      I sometimes wonder if this is what is happening with some of the news articles I read. There are so many grammar and spelling mistakes that I notice (that drive me nuts as a teacher) and I wonder if there is any editing done anymore, but if the article is dictated/transcribed and not checked over well, then that could account for the errors.
      Just a thought…

  2. Hi Marlis,
    One problem I also faced was not being able to have paragraphs when using dictation. When I used my iPhone’s built-in dictation function, I had to say when to add a period or a comma, which I often forgot to. I think your transcription was in general very well done, did you have to say punctuation marks? I think you described it really well when you said that oral storytelling has a more personal element to it. There’s less time to consider what is right and wrong, and the emotions that come out of oral storytelling I feel like is more authentic in that sense.

    Emily Chen

    • Hi Emily – thank you for your comment. I use the dictation feature quite often when texting, so I’m used to adding punctuation. I have to say the intended punctuation when I want it, so it sounds quite funny to hear me speak into it. A text to my husband might sound something like this…
      “Hi there comma I hope that your day is going well period Are you able to stop by Save-on and pick up some groceries on the way home question mark”
      It generally works well once you get used to it, but then if I actually wanted to use the word ‘comma’ or ‘period’, I’m not sure if I’d be able to override the punctuation feature.
      Proofreading is still quite important because despite how good the app is, mistakes still happen – damn autocorrect, lol.

Leave a Reply

Your email address will not be published. Required fields are marked *

Spam prevention powered by Akismet