Search Here ….

Tags

Academic integrity AI AI Detectors Artificial Intelligence Asia School of Business ChatGPT ChatGPT Malaysia ChatGPT Plus Education Entrepreneurs and Startups Malaysia Entrepreneurship Gardens Google GPTZero IGCSE Malaysia Malls Maps72 Microsoft Midjourney MIT MrMoney TV OpenAI Pavilion Plagiarism Plagiarism checkers Poe Quora Sam Altman Software development Technology Turnitin Udemy Writing
Uncategorized

Some thoughts about ASR

Victor TanPost a Comment

What you’re reading this the result of dictation from an M2 Max Macbook Pro, to which I upgraded after some time of pretty much just realizing that I needed a new computer.

I don’t know how good it is going to be, but I have expectations that it will be a little better compared to what we have on the iPhone and other devices that we can use as part of the Apple system. So far, it seems there are some issues with the system because it recognizes some words incorrectly – partly that could be because I am pronouncing those words in a way that is not really concordant with what the algorithm is able to imitate. Hence, it ends up transcribing the wrong things, because what it hears is, in fact, something incomprehensible.

When it comes to automatic speech recognition, the algorithms that process our speech face problems when the expected input is ambiguous and can possibly match several possible outputs. The automatic speech recognition function is, after all, based on making predictions, based on the likely input that would be expected from a user over time. In the event that the prediction made by the algorithm is incorrect, the software may be penalized as a result of the input not matching the prediction. The ultimate goal is to have a scenario where every prediction generated by the model, as the person speaks, matches what the person intended to say with the least possible corrections required.

Let’s start with the voice’s pitch and ambient noise.

For an algorithm to receive the correct input, the sound it receives should match the specific waveforms used in the training dataset that guided the development of the automatic speech recognition system. If there is a deviation in the sound pattern, either due to ambient noise or tone of voice, the generated text can be inaccurate because the system receives problematic inputs in the first place. This is natural, because if you feed garbage in, you will naturally receive garbage out. There is a sort of equivalent exchange at play.

Let’s now talk about something different altogether – the device’s processing power. I’m not sure if this is a factor with the automatic speech recognition system on Apple devices, but I suspect it is. Each device needs to perform complex calculations that allow it to make predictions at a relatively high rate of 100 to 130 words per minute. If you look into the memory usage of the computer as it performs this process, you may see that the memory does not get consumed at a high degree – that is something I plan to test in the coming days.

The last possibility is that there is a problem with the algorithm itself in recognizing certain patterns of waves. There can be some variation depending on the quality of the input, but it’s also possible that the algorithm used to process the data can make mistakes on occasion. I’m confident that Apple is making strides to improve the output quality of its algorithms, and for that reason, I am optimistic about the improvements they can bring about.

I give this much thought because it is one of the most important aspects of any generative artificial intelligence system. These systems rely on good input into the algorithms, and speech recognition systems are extremely important as sources of input. As we interact with natural language on a daily basis, which is often faster than typing or pressing buttons on our devices, I believe that accurate dictation and procedural proofreading of our daily writing will lead us to a new era of AI.

I can’t wait to see what the future holds, particularly as we approach the end of 2023 with developments like iOS 17, AI generators, and all the different forms of technology that are becoming more prevalent. Time passes, and it is somewhat sad to think that we are coming closer to the end of our lives before these things come to fruition – still, the show is not over until it is over, and I can’t wait to see what is going to come!

Leave A Comment

Recommended Posts

The Body is the Hardware, The Mind is the Software

The analogy was interesting when I heard it first, and it remains interesting now because it resonates with me on at least a couple of different levels. Our bodies, the physical parts of us, are basically analogous to the hardware of a computer, running along with different parts here and there – upgradable, we can improve them by increasing the quality of the resources that go into them; improvable through good maintenance, we can exercise, sleep well, and do all sorts of other things to improve the hygiene on that front. Our minds, on the other hand, are the software – the programming that decides how we interact, think, solve problems in specific situations; the algorithms and little decisions that decide how we react to different scenarios and confronting different situations, whether it comes to talking to girls, investing, selling, marketing, or doing business with others. It is nice to think that the mind is upgradeable, and that somehow you can improve yourself through an act of willpower by learning certain things. Through sitting down and unlocking the secrets of the universe one after another, through a mixture of magic and also destiny. But who’s to say exactly how that should happen? Sorry, that’s a silly question. The answer is that it’s you.

A Small Change of Perception

I began this morning with the headline “How Kamala Harris Burned Through $1.5 Billion in 15 Weeks”, on NYT. It was an interesting head to a week of what was for me listening to, understanding, and better reckoning the world after Donald Trump was elected 47th President of the United States, and the first of many headlines I’d seen about this on New York Times. Some might view this as evidence that the media is cleaving towards the Trump administration as the chickens fall in line and loyalty becomes a Sine Qua Non in the era of an evil empire – but I think a little differently, because I feel like it’s teaching me something about reality. Look in the comments, and you will see how people have responded – people saying that the presidency is “deeply unserious”, highlighting any number of things that they disagree with even as they say that NYT’s “focus” is wrong, that Kamala “tried to save democracy”, and everything in between. If I really think about it, all of these seem about as valid as saying that Trump is secretly a genetically modified orange with a toupee made of cheese.  The entire idea of NYT is that it’s one of the most respected voices in journalism, that alongside publications like the Washington Post, it defines the Overton Window – the space of ideas that are acceptable to the public at any given point of time.  To the extent therefore that NYT’s function is valid for this purpose, I’m more likely to say that these critics are the ones who don’t make sense – That the calls against that validity are the true measure of what doesn’t make sense. I’ve often heard this idea that in fact a Trump presidency might be a situation where the inmates are running the asylum, but upon further inspection, I’m no […]

Perfectionism to eliminate

…And another has come. We are progressively moving towards the end of the year with each new beginning. This is I believe the 46th week of the year out of 52, and it’s leading towards the end of the year; Donald Trump is now president, filling up his cabinet with appointee after appointee as people contemplate things; you might believe that we’re at the start of something HUGE, as Donald might call it, world-shaking, incredible. But I think while that’s good, it’s good to look at something that I’ve wanted to get rid of for quite a while: Perfectionism. I am a victim of it, and I can’t deny that it follows me everywhere, making me question myself and whether what I’m putting out into the internet is either good or worth it – I second guess myself frequently, taking down blog posts that I think aren’t great or that aren’t well worded, thinking that perhaps I should rewrite or otherwise. I think that this is a very negative behavior, because frankly I don’t really care too much about what people think and secondly, it doesn’t really matter what they think – at least in relation to how I think about myself. So I would like to eliminate, therefore, the perfectionism that makes me rewrite things, redraft things, take wayyy too much time to release things. This is the next thing to change, and it’s a good thing to shift it in this year of 2024 – even if it is the only lesson that I will have learned by the end of this year, I think that it will have been a worthwhile one. Here’s to the next!

Creation

On an empty page, the pencil traces the dotted line, the circle, the shape.  From the movement of the pencil, a million universes appear, timelines splitting into their multiple component parts in a universe of endless possibility as millions more appear, each one a multiverse of possibilities as the pencil moves, tracing by movement, through which, across billions of possible environments, worlds, universes, colors, shapes, and relations. Look up from the page and perhaps you may see the created universe – Breathe in and you may appreciate its harmonies, the unity of physical constraints, of physical laws interweaving to make existence possible.  But is that universe truly greater?  Look down at the once empty page, no longer so empty. I claim that if you look further, there you will see it: Here lies the immortal beginning of every endeavor, the step not taken – a journey not yet made of a thousand miles as yet untraveled that you can begin, where you are, with a single stroke of your pen. Here, then, is the possibility of a universe even greater and even more intricate than you may have ever known — Whether you can reach that universe or not? That is a separate question — and none but experience can teach you its answer.

Your Teacher’s Thoughts towards the person he likes (but he doesn’t know?)

The feeling of love for someone is not something that you just go right ahead and deny. I don’t think it’s something that you should be shy about: That you like a person. Somehow after the years have come to pass it becomes true that somehow or another your ego doesn’t really take that much occupancy. You can admit honestly that you like a person even if you realize that there is no expectation of a return. On my part, I don’t know; even as a teacher, I really like a couple of different people here and there, but I realize clearly that there might not be really an expectation of a return, and that’s okay, that’s just how I am – it might be strange to think about, but even your teacher might have emotions, and so too might the rest of the world. It is a little unconventional to reveal your emotions, especially in a world like this, but to the person I do like, I like you, but I realize that I should not put you down, I should not bind you, I should not stop you from being who you want to be. If we come together, it is because somehow or another, through the millions of possible pathways, and somehow through the conversations, we liked each other, and that is enough for me and is an act of fortune, not of planning or otherwise. In the past, I would’ve been afraid of saying that I like a person or I wouldn’t have been so honest with my emotions. Nowadays, I don’t know if it’s because I’m old now, but I think it’s okay to say that I like people and I’m not too afraid of saying that I do because that’s just what it really is – an expression of emotion and a reality that I […]

The things we like but are not good at.

In this world, as we pass through, we may realize that there are some activities that we deeply and truly love – little skills, hobbies, and occupations that pique our minds, hearts, and souls when we participate in them as an act of pleasure.  As we pass through the tides of time, though, almost inevitably we come to realize that simply because someone we enjoy something, that doesn’t mean that we are going to be good at it. In fact, that’s an understatement.  Why are we talking about good when actually we can be horribly, devastatingly, and world-changingly catastrophic at it?  Here the realization inevitably comes, almost as if it were the common heritage of humanity: Just because you like something, that does not mean that you will be good at it.  One might argue that a true passion is such that even if one isn’t good at something, that the passion should stay.  Even if you are a horrible dancer, that does not mean that you should despise dancing.  The words of an eternal Malay proverb come to mind, “Tidak tahu menari, memarahkan lantai.” They resonate through the core of our beings and remind us:  If you dance horribly, that does not mean that you should blame the floor.  In other words, our lack of skill is no justification for our preferences, which are shown superficial if being bad at them is our grounds for casting them away.  After all, are we not like the fox, that declared the grapes sour, purely because we could not reach them?  In a way, this may be true, but a reality is that in this world, skills are not necessarily their own reward, and imagining that they are is to neglect the realities of our universe in lieu of something all too idealistic, rarefied, and divorced from both the world and the way […]