Speech to Text… Take Three

I grabbed a new tool. I set it up and tested it out on the drive home. I thought about just posting the raw words, but it’s a rambling mess. What I did get though, was around 1,800 words of raw, rambling mess – so this looks like it might help me keep moving posts like this forward more efficiently. There will be some edit needs, but this might actually work this time.

So here we go for test number three! I tried out a different app yesterday, rambled on mercilessly for probably 15 or 20 minutes. When I went back to look at the text from that particular app – it had stopped recording as soon as the screen shut off. That rambling mess –  almost all of it went away. I hope to get the little piece of that posted in the near future. That’s  the point of this. I want to be able to put together some thoughts in a more timely manner get things out for blog posts. Move things forward. There are a lot of times when the moment I sit down and I think to myself, “Okay, here it is. I’m going to… I’m going to write!” and I end up knocking out a blog post when what I should be doing is working on one of my stories.

That was as far as I got before this thing errored out.

I think that’s going to be a big problem. I’m trying to do this while I’m driving. It allows me to use the drive time usefully.

This thing has errored out like twice while I’m driving and that’s really sort of the point. It needs to be working while I’m driving. I need to not pay attention to the app and focus on driving. That is the primary thing I should be doing. The app should allow me to use my drive time in a more productive manner.

I mentioned the drive time to a friend at a convention recently and he suggested it. He suggested a different app, but that different app uses AI. I don’t know if this one uses AI or not cuz maybe it doesn’t. I will check it out. I am not a fan of using AI. I had a whole rambling thing about AI yesterday and I hope that this one doesn’t go away. If I see the error message pop up out of the corner of my eye I’ll just stop and then we’ll come back to it later. Until then I’ll just keep going.

So – drive time. I want to use this drive time productively. I want to use it for getting thoughts out. Stuff that I can use for quick edits to send into a blog post. If I can get the blog post put together quickly, that saves me time. I do want blogging to be a thing that I continue to do. I do want to continue. I’ve done it for years, but have fallen off a lot after recent events. I’ll start working on stuff and fall off because time, then I’ll start working on stuff again. I put forth a lot of effort in just starting and stopping. I’m throwing opinions out there and keeping the ‘publicity’ going, but what I should be doing is writing more of my stories. I need to get my work out there, so we’re going to start with the blog post and see if I can move forward faster. 

Moving to the stories as far as using this speech to text thing won’t be as easy. The first thing that I noticed after yesterday’s attempt is that speaking for number of minutes continuously not an easy thing. Particularly if it’s on scripted. I was all over the place. I was throwing thoughts at this thing yesterday from all over. I should probably be thankful most of it was lost, but even today I can tell already – this is going to be super stream of consciousness.

 I don’t know, or I haven’t quite figured that part out yet. The most important thing is to make sure it works first. I will just chatter on as I’m on my drive home and we’ll see when we come up. The app says ‘unlimited length’ for posts and I’ll put it to the test.

Another part is I want to be traveling with purpose. So many topics have slid past just based on my ability to get to the computer and type. I want what I have to say to have some kind of meaning as well. I say that because I’ve discovered that I’m still very upset about something that happened (many years ago) with a local news source. I went on a rant yesterday about this and I got caught up calling it a newspaper and describing a blog in column inches… it wasn’t great. The incident was that this news source had a columnist that they paid – yes, an actual live human – and actual live human was in a columnist position getting paid and you are potentially blocking somebody else out of that position doing actual work. What bothered me? This individual went on at length saying, “well… I’ve got nothing.”

I was angry then and I’m still annoyed now. IF you don’t have anything to say anymore, get out of the way! That person got paid! It just… it bothered me. I saw a story recently that said she was still that person until she recently retired. I shouldn’t be so upset about it. I suppose that’s pretty much any channel these days. If you turn on any media source there are a whole lot of people with nothing to say.

Having this time should help me add words.I did some rough math in my head. Even if I’m only throwing down… keep it to round numbers here… I manage a thousand words three times per week when I’m driving back from the office. That’s 3,000 words that I didn’t have before. 3,000 words a week for a month, there’s another 12,000. Hopefully keeping things moving. Trying to do the math in my head that’s like 72,000 words or so in half a year. A pretty good chunk of a novel right there. With edits and formatting, I think I could put together a book in a year. That’s if it’s only the words that I use while I’m using the speech to text. That sort of thing is a huge boost, so fingers crossed. 

Doing this will mean, vocal chords warming up. It takes effort, it takes time, and if I’m going to do anything with a story I can’t just off the cuff throw Story Stuff down like this and have it be successful. Side note – talking about hundreds or thousands of words is great, but that’s if they’re good words. There’s no guarantee that the words will be worthwhile at all. There’s the very real chance that it’ll be half that amount just because most of it, the majority of it is simply garbage. I’ll be interested to see how that was typed. I pronounced garbage with a bit of an accent. A person who’s speaking doesn’t always translate directly to text. What you’re saying, the emphasis, the hesitations… those sorts of things don’t automatically come through.

I have always struggled with moving a story told to the written word. I think that’s part of the explanation of why my writing is so slow. If I’m talking to somebody and we’re having a conversation and I have emphasis and I have sound effects and accent and I can tell a really good story. That doesn’t always translate directly to text. While I may be a decent storyteller that doesn’t make me a good writer. It’s a really hard thing to translate from one to the other.

So I’m testing out my stream of Consciousness here to see how it looks, how it works. I’m certain that I repeated myself probably three or four times within the context of the last 10 or 15 minutes but that’s the other piece of this. When you don’t have a plan, when you don’t have an outline, when you don’t have a goal in mind, then you just go you ramble and sometimes you get repetitive and sometimes end up with the similar word creeping into your text over and over and over again. I might need to find and replace, or just a word search in general to find that word that you’ve used again and again and amusingly enough again should probably be one of those words.

Having said all of that before this, I do intend to research whether or not this tool uses AI. Part of that is I am not a fan of AI. My background in science fiction has shown AI as the bad guy so much, for so long, it’s difficult to escape from that mindset for me. I have a minimal understanding of large language model training and that sort of thing. It’s certainly not enough to say I understand it. All I could really say is that I’m vaguely conversational about it. It’s emotional for me. It’s the feeling portion of it I’m not comfortable with. If I don’t feel comfortable with it then I’m not going to want to use it. There’s also a part of me that wants to do the background research because all of the news lately shows the amount of resources that AI is consuming in terms of power and computer chips and water resources. All of that sort of thing along with the massive number of data centers that are being put forward in planning processes right now… that feels like the sort of thing that could very easily get out of hand. Far too easy using it and not being aware of how it’s going to affect our environment.

For my purposes, tools that I have seen so far that companies have put forward for AI do a lot of things very poorly, very quickly. I can get you a lot of wrong answers in a heartbeat. If we are going to depend on them then we definitely want to make sure that we are prepared. Correct answers for things that we want to be able to have a level of confidence in. When we say “AI did this and I’m pretty sure it’s correct” it’s not always enough. I don’t want to say “hey AI did this and I never checked it out” and then find out at some point later it absolutely didn’t work, was wrong, and now the stuff that I’m working with is wrong and that, that really bothers me. Do I think AI is going away? Absolutely not. There are too many companies that have invested too much money. They’re putting AI in places that it absolutely does not belong. Eventually the market will find an appropriate level and straighten itself out. When we know the appropriate tools, the appropriate use cases of those tools, then we can deal with it. It does NOT need to be in everything. In fact it bothers me that it has been forced into my email account already. I don’t want it there. I don’t need it there. I don’t need some computer telling me ‘hey I can help you write that better’. No, you really can’t. You’re going to use generic terminology that doesn’t fit my voice. My stories, my voice. That matters. Not enough people get that.

There it is. My massive, rambly, on my way home piece. I am now actually within probably 5 or 10 minutes of the house and I have just let all of this go~

And it errored out again. A fine point to wrap up at.

Far more editing time than I wanted (close to an hour). More words than I thought (final count just over 1,900). A successful test. More rambly words to follow!

Speech to Text – Take Two

Just imagine the little movie clapper thing making that snapping sound… because imagination is about as far as I got. Here’s the text:

As I am on the road home for testing this out for the first time and I have no idea well I have no idea number one how long I’ll be able to just speak at this thing number two I have no idea what exactly I’m going to speak about just trying to test it for the sake of testing it and number three I suck an indexing and I didn’t have a third thing so now comes the real test because the screen is about to shut off and I’m not sure where exactly the whole thing ends or starts or how long a pause I can have if I don’t have a actually stop doing this see if it actually picked up all these words that’s it for now post the results of this test up to my blog and then we’ll try it again later

That is the rambling result of more than twenty minutes of talking. Turns out the mic shuts off if the screen goes off in that app. I have to consider that a complete fail. I have no use for a tool meant to keep my hands free if I have to keep paying attention to it and poking it every couple of seconds. It didn’t keep up well with my speaking pace, nor did it catch everything.

I’ll look up and try another one here over the next day or so. This one is getting uninstalled.

Tool Testing

I recently changed firms at my day job. The new place is a much further drive than I have had in some time.

I was talking with a friend about my current place of employment that I when I have to go to the office, I need to drive more than I have in a long time. It seems to be a real waste of time. To have this 45 minute period each way that I can’t really do anything with is wasteful. To be fair, I have listened to more news and more music, but there are days when I need more time and that seemed like a usable spot. He suggested that what I needed was a way to talk to notes – use a voice to text transcription tool of some sort. I could use that time driving productively by taking notes from my train of thought, then quickly editing the notes so that I could keep my stories and my blog moving along at a better pace.

Even if I only got a few hundred words, that would be a few hundred more than I had before.

This is not the fist time I’ve tried a tool like this – but this is the first time I’ve had specific time when I couldn’t do other things, or type at all. This was my first attempt with an app called Voice Notebook.

As with any voice to text app I’ve encountered in the past, the editing it the real challenge. I have managed to move the notes to a cloud drive and clean them up. We’ll see how this goes when I need to use it in the car (I cheated and tested this one while I was sitting at my desk in the office).

Toolbox Fallacy

It’s still weird to me to have my words translated onto a page by way of a microphone. This sort of thing is going to take some getting used to. I need to learn to speak louder and more clearly when I do these things. Having said that, I hope to be able to create many more words very quickly over the next few weeks.

Oddly, when I’m speaking into the microphone like this, the flow of consciousness helps me to organize thoughts. Sometimes, however, one thing that it does is make other thoughts and other patterns of memory more evident. I was telling somebody the other day about the toolbox fallacy. Is this a philosophy?  Perhaps it’s a school of thought? Is that the same thing?

Anyway, the thought here is that people who are creators, in whatever form they create,  simply create. The tools are just that, tools. If I want to say that I am a writer, I should write.This should mean that I write without regard to what tools are available to me. If I need to use a marker and a napkin, or a piece of cardboard at work and a pencil, whatever it is, if I have an idea I should be writing.  Editing, compiling, composing all of those things come after. The most important piece is the actual creation. Get words on a page. If the page is blank, there’s nothing to react to.

The caveat here is, I need the time in my schedule to actually sit and make these things  happen. I’m hoping as we move forward through the end of this year that I will be able to make more time for my creative pursuits. I really do want to finish the first full novel that I have been writing for a very long time. I have world building, I have characters, I have story arc, I have villains, I have plot twists, but what I don’t have are words on a page. Hell, I’ve got as much “marketing tool writing” as I do actual, written story. Can’t buy it if it’s not actually there.

 So, I’m going to take this stream of consciousness as a starting point, remembering the toolbox fallacy, and try to create more and get these words out of my head faster so that I have something on a page that I can then go and edit and say yes I have completed this thing.

 I’ve stated this before, but I’m hopeful that perhaps, this time it will stick. Only time will tell.

Check out this video – it’s part of what stuck in my head about making my work real:

New / Old Tool

I’m going to test out an old/new tool again. I had tried using voice to text previously and one of the things that slowed me down was always the amount of time that it took to edit these things.  There’s no way to actually say, “hey that’s the end of a sentence”. The Voice to Text tool will actually spell out the word. Punctuation is the bane of voice to text. Pauses are not recognized.

Interestingly enough I think someone complained about that and if you pause long enough you can actually say. and get that actual result (see how that worked right there?)

I’m going to try this again. I can speak faster than I can type. The real question is how much editing do I need to do when all is said and done. I think it will be an interesting experiment. It will force me to look at the words that I create in a much more conversational format as well as giving them something of a flow that they might not have otherwise.

Maybe I get these stories, these words, these things to come crashing together faster and hope the editing process is easy. Maybe I get results that amount to a stream of consciousness that don’t really work out as good stories. Hopefully one of the things that I can do is make this stream of consciousness thing work for me and use my ability to tell a good story to actually create a good story on a page. Typing in this way, which isn’t really typing, I suppose I should at least be able to pull together words faster and more completely than I could do in a traditional typing manner. If I’m lucky I will be able to produce more, and do it more quickly.

I think this is going to be at least, in part, something that I’m going to have to learn and I’m going to have to spend time working with. Based on an articleI read recently, it was suggested that what I really need to do is learn to live with an absolutely terrible first draft. I tend to want my story to come out as close to finished as it can right from the start. I don’t like the editing process. Not that I think anyone likes the editing process, the editing process sucks.

This is the test and we’ll see if my production speed goes up. One of the things I really need to get better at is my speed. There are so many things going on that I really need to make the best use of my time and use tools that maybe I wouldn’t have thought were positive or useful or even for me. We’ll see how it goes.

Time to edit this post because I can already see a giant mess.

Do you have any tools that are your favorites? What tools help your productivity?