Separate names with a comma.
Discussion in 'iPad 3 Forum' started by pego99, Mar 19, 2012.
So it translates as you speak?
I believe you would have to get an app to do translations. Several are available, but I do not if they are adapted to the dictation function of the new iPad.
I am not interested in translation just typing in real time as I speak from my voice in english. I thought I saw a apple ad showing this but I am not sure.
I actually find the voice to text feature works quite well, and have written a page of text, and was surprised that I only had to edit about 4 words which it got wrong.
Just tap the microphone icon when the keyboard pops up... simples. I find a lot depends on the words used.. sometimes it will make quite a few mistakes, but overall it does a pretty good job, though obviously one has to acknowledge that utils like this will always have challenges. A lot depends how clearly one speaks, and how the words are actually spelt & pronounced. I think it's something I may use fairly frequently. I used it a few minutes ago to do a text list in Notes and the results were faster than I could type it, though I did have to edit 5 words afterwards.
Mine does not type in real time. When I touch the microphone button and speak I have to push the microphone button again to have it type what I said. I saw a demo where the typing continues as you are speaking without having to push the microphone again. Has anyone seen this mode?
I don't know what video you saw...it works on whatever you say between pressing the microphone button in the keyboard. Real-time dictation is a lot harder because it has to listen and process all at the same time...and the resources to make that happen are bounds greater, depending on what is being said and how fast it is said. By just operating on what is between the button presses, the process is never allowed to get out of hand...performance can be guaranteed that way. Your way it cannot be.
I see the same thing. Seems like the text should appear as you speak, not all at once. Strange...
Real time diction requires all the software resources to be on the device, and it's a very CPU intensive application. I doubt the iPad has the resources to do this, and if it did it would be a much bigger app.
Most of the current diction apps send sound files over the internet to powerful servers to be interpreted. This is why you don't get immediate translations. This is much more efficient, and the longer the sample the better. Partly it is to keep the amount of data down, but mostly because the servers need a large enough sample to detect accent and use context in the translation.
Accurate single word translation is almost impossible when dealing with a wide range of people and accents. Computer based applications get around this by learning your voice over a period of time or with a training session. So far no one has set up server based apps to do the same (as far as I know). If they have, it will involve logging in to the service in some manner.