Say something in Cornish is great, but not quite so good for listening to the phrases and seeing if you understand them. To that end I’ve written a small program. You listen to a random phrase, and then type your translation and are told if you got it right or not. You can also choose to see the text of a random phrase so you can say it out loud before listening to what it should sound like (so like a random chunk of SSi Cornish).
The program is written in Java so will run on Windows, Mac, or Linux. You can choose what to listen to, such as mostly lesson 2 but with a few lesson one phrases thrown in. Currently I have all of lesson one phrases and the first part of lesson two, but will add more daily as I progress. However you can add your own phrases yourself, you just need to cut them out of the downloaded lesson .mp3 file. There is a free program on the internet called Audacity that makes this easy, and there are others too. Because of this, this program can be used with any of the SSi courses.
I can’t upload the program here, as only images can be uploaded, but if you let me know you want a copy, you can have it for free, along with instructions on how to use it. I will also send you the extra sound files as I make them each day (at the speed of my learning).
All of lesson two and some of lesson three now included. The following updates are now included.
If you give a valid answer, but not the one the program is expecting (Such as “I need to do it” instead of “I have to do it”), you can now tell the program that the last answer was correct. It will remember this for the future. These aliases as stored in a text file, so you can edit them yourself manually if you want.
If you give the wrong reply, the phrase will be asked again when you have done all the other phrases.
If you make a typing error on what would otherwise be the correct answer, you can say that it was correct but not to add what you typed to the aliases. This way the phrase will not be asked again at the end.
There is more control over what phrases are given. You can now say: A few of lesson one, all of lesson two and three, but for lesson two only give one of the two versions of a phrase. This way, for lesson three you will at some stage hear both Pol’s and Julia’s version of a phrase, but for lesson two you will only hear one version of a phrase (It’s random as to which one for each phrase)
All of lesson three is now included. The following update is now included:
After completing the read and say phrases, you will be told what percentage of them were answered correctly within the time gap in the lesson (for each phrase) before Julia starts to speak. You can also request the stats when you are part way through answering phrases.
Is the software FOSS? If so, can I peek at the source code?
I’m not a Java dev (shell, Python, Ruby, Go), but I’d love to see how you implemented this.
Sure. Send me a message with your email address and I’ll send you a copy and I’ll also add a copy of the source code. Give me a couple of days because I want to write a few notes. Are you on Windows? There is a tiny .bat file to invoke it. If you are on Linux or Mac, I’ll write the equivalent shell script.
I’ll keep hold of the source and distribute it myself. The reason for this is that with the SSI lessons, you have to register before you can listen to lesson 2 onward. Since the distribution contains clips of the lessons, I can check that whoever wants it is registered here.
I don’t mind you seeing the source code though. I’ll stick a copyright in comment just to cover myself. .
Steve,
I’ve been enrolled on Saysomethingin.com (Cornish). for some time. My a dhallathes ow tyski Kernewek nans yw vledhen. Splann yw, ow tyski Kernewek omma yn Kyoto, gans koweth in Yokohama, warlinen.
If the offer is still open, please send me the software. Much obliged.
Peter Trebilco trebilco54@gmail.com
