The development recently with major languages (LLMS). Many of the focus is answering a question you can do with the text-based models based on the text, or VlMs), where you can file photos.
However, there is a great extent to a tonoon for the past few years: Audio. The model models are able to write (speech -> Text), integration of speech (text – and talk-to-talk, where you meet audio both in and out.
In this article, I will discuss how I use the development within a sound model space so that I have my benefit, and I have a very efficient system.
Motive
My first motivation to write this article that I am always looking for ways to be effective. After using the mobile phone app temporarily, I get their writing option (Microphone icon on the right on the user installation field). I used the text and I soon realized that this was written in comparison to others I had used before, such as the built-in Apple.
Openai's Openaai is almost always holding my words, with very few mistakes. Or I use ordinary words, for example, acronyms related to computer science, still knowing what I mean.
This text was only available in the ChatGPT app. However, I know that Openai has API ENDPOINT for their back model, ie (maybe) the same model they use to write text to the app. That way I wanted to set this model to my Mac that it was available for activist.
(I know there are applications such as Machispherers available, but I wanted to develop a completely free solution, without the cost of the APIs themselves)
Requirements
- Alfred (I will be using Alfred in Mac to cause some documents. However, some ways in this and exist. Usually, you need a way to create your documents.
Entercourishing
The main advantage of using this text can enter the words on your computer as soon as possible. If I type as soon as possible on my computer, I can't even reach 100 names in a minute, and when I type in the vine, I have to focus. However, the middle speed is minimum of 110In accordance with This is the gearCLEA.
This means that you can be very effective if you can speak your words in writing, instead of typing them on the keyboard.
I think this is closely related after the increase in large-language models such as chatGPT. He spends more time releasing language models, for example, asking questions to discuss, or move the cursor to use a feature, or fix the bug. Therefore, the use of English language is more than ever now, compared to the use of planning languages such as Python directly.
QAPHELA: Of course, you will still write a lot of code, but I spend a lot of time, for example, for the extent of the English, where the text is used to save a lot of time.
Illustration
However, other texts use the text. One of the priorities is that many times, you don't want to talk aloud when the program. You may be sitting at the airport (as I am when writing this article), or even in your office. When you are in such circumstances, you probably want to interfere with those around you by talking aloud. However, if you are sitting in the home office, this is not a problem.
The other is not true that a minor increase may not be as soon as possible. Consider: As long as you just want to write a single sentence, it will, in many cases, hurry the storm by hand. This is because of delay in the beginning, standing, and sound writing about the text. The API call takes less time, and the immediate short, a large fraction of the time you should spend and wait for the answer.
How to Use
You can see the code I used in this article in my Gitub. However, and you need to install Hotkeys to use the Scriptures.
First, you should:
- Clone The GitHub Repossitory:
git clone - Create visible environment called .The trust Then enter the required packages:
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt- Find the Opelai API key. You can do that:
- To go to Openai API view, enter / create profile
- Go to your profile, and API Key
- Create a new key. Remember to copy key, as you will be able to see you again
Scriptures from GitUB repository works:
- Start_Recording.SH – Starts to record your voice. First when you use this, it will ask permission to use the microphone
- Stop_Recording.SH – sends a sign sign in the text to stop recording. Then you send a recorded sound to unlock the Onvion Tradrition. In addition, you add a text text to your chat board and paste the text if you have a text field on your selected PC
The entire storage is available for the MIT license.
Fly
You can get Alfred's work movement on GitHub Repository here: Writing.AllredworkFlow.
Here is how I put the flow of Alfred's work:
You can just download it and add it to your Alfred.
Also, remember to have a terminal window to open at any time you want to use this sker, as you activate the Python text from terminal. I had to do this in this way because when the script is done directly from Alfred, I find permission issues. The first time you drive the text, you must be asked to provide your signal access to the microphone, which you should agree.
Charge
Important consideration When using APIs such as Openai Whemper are the cost of Apply API. I can check the cost of using the highest openai model. As regularly, the cost is completely dependent on how much the model has been using the model. I can say that I use the model up to 25 times a day, up to 150 words, and the cost is below one dollar a day.
This means, however, that if you use a model too much, you can see the costs up to 30 dollars a month, which is more. However, I think it is important to note the savings you have from the model. If each model use saves 30 seconds, and using 20 times a day, you have just finally lasted ten minutes of your day. Personally, I'm willing to pay one dollar to save ten minutes of my day, do the work (write on my keyboard), that doesn't give me any advantage. If any, use your keyboard can affect the high risk of injury as carpal tunnel syndrome. Through the model so it's worth it.
Store
In this article, I began to discuss great progress between language models in the last few years. This helped us create powerful Chatbots, save us a great time. However, with the development of language models, we have seen developments in voice models. Writing using Openai Wrhshsws is completely close (from personal experience), making it a powerful tool that you can use to write words on your computer successfully. I discuss the benefits of the benefits of using Openai Whishere on your PC, and I too got step by step how you can use on your computer.
