Scratch

Since tone length means increasing or decreasing the amount of time, that would mean that the difference would have to be great enough for the receiving project to interpret it and decode its length (which, again, needs some form of delaying and timing). That means at least 0.3 seconds for the quick tone and (counting an interval of about .3 seconds) a minimum of 0.6 for the long tone. A string of text the length of "hello" (not case sensitive) could mean about 15 seconds!!!

Volume is overall the best option I guess. using .1 second impulses, the volume would vary from 30 (or even 0!) to 100, the difference being the boolean "loud?" or not "loud?" to speed up the recognition (only slightly, but could reduce errors).

The same string "hello" would then be of only 3 seconds. A considerable difference.

To solve the surround sounds problem, and even the mike volume, you could include a volume test calibration system.

What do you think? 

Last edited by LS97 (2011-01-05 13:49:14)

Maybe a "wait until tone is played, then adjust the distance between the speaker and mic and the speaker volume until the arrow is in the green zone"

Message with an arrow that changes X position to loudness, then play a mid-volume tone that should be 50 and wait until it is 50 with say a tolerence of +-3 

I'll help, but I can only help with "idea-ing".
My house is too loud to test these kinds of things. 

Forgive me for not knowing, but what is a "jack to jack" cable? Knowing the names for things isn't really my strength. 

That's cool.
But I don't have that. 

Here's a nice pic!


and here's a link to just how cheap they can be online: link

Last edited by sparks (2011-01-05 17:56:40)

I didn't realise we needed a name for it, but now you mention it, it would probably help if we have something to call our finished software    Maybe leave it for now and see if a name suddenly sticks later on 

Btw, there's a link and a pic in the post above yours for you 

hmmmm, you should have a speaker output on the back of the laptop or on the side somewhere, though I'm not sure where as I don't own one

Ah...the sounds of audio encoded data streaming through the air!  It brings back fond memories of modems and phone couplers.  A cool idea from a "can it be done" viewpoint 

Having discussed this some more with a friend, he points out that as in morse code, it's a good idea to specify the most used characters to the shortest tone periods in order to speed up messaging. Unusual characters can be reserved for longer, less needed tones! So I'm going to start experimenting a bit and also, if someone could find a list of ascii characters in order of usage?

Really? You can buy them sooo cheap! You could still have a look for that list 

It's easy to find one for letters, but all characters might be a bit harder to locate 

results

Instrument: Electric piano 1 (5)
note duration: 0.5 beats
sending program: Scratch. (windows XP on desktop PC)
recieving program: Scratch (turbo speed on)(Windows 7 on a laptop)
connection type: Jack to jack cable. (1m)

Results:

volume                    maximum loudness registered.

1) ----------- 0
2) ----------- 0
3) ----------- 0
4) ----------- 0
5) ----------- 0
6) ----------- 1
7) ----------- 1
8) ----------- 1
9) ----------- 2
10) ----------- 2
11) ----------- 3
12) ----------- 3
13) ----------- 4
14) ----------- 5
15) ----------- 6
16) ----------- 6
17) ----------- 8
18) ----------- 8
19) ----------- 9
20) ----------- 10
21) ----------- 12
22) ----------- 13
23) ----------- 14
24) ----------- 15
25) ----------- 17
26) ----------- 18
27) ----------- 19
28) ----------- 21
29) ----------- 23
30) ----------- 24
31) ----------- 25
32) ----------- 28
33) ----------- 29
34) ----------- 31
35) ----------- 32
36) ----------- 35
37) ----------- 37
38) ----------- 38
39) ----------- 41
40) ----------- 43
41) ----------- 45
42) ----------- 47
43) ----------- 50
44) ----------- 53
45) ----------- 54
46) ----------- 56
47) ----------- 60
48) ----------- 62
49) ----------- 64
50) ----------- 69
51) ----------- 70
52) ----------- 74
53) ----------- 75
54) ----------- 80
55) ----------- 82
56) ----------- 85
57) ----------- 87
58) ----------- 92
59) ----------- 95
60) ----------- 96
61) ----------- 98
62) ----------- 98
63) ----------- 98
64) ----------- 99
65) ----------- 99
66) ----------- 99
67) ----------- 99
68) ----------- 99
69) ----------- 99
70) ----------- 99
71) ----------- 99
72) ----------- 99
73) ----------- 99
74) ----------- 99
75) ----------- 99
76) ----------- 99
77) ----------- 99
78) ----------- 99
79) ----------- 99
80) ----------- 99
81) ----------- 99
82) ----------- 99
83) ----------- 99
84) ----------- 99
85) ----------- 99
86) ----------- 99
87) ----------- 99
88) ----------- 99
89) ----------- 99
90) ----------- 98
91) ----------- 98
92) ----------- 98
93) ----------- 98
94) ----------- 98
95) ----------- 98
96) ----------- 98
97) ----------- 98
98) ----------- 98
99) ----------- 98
100) ----------- 98

Notes: As you can see, there is no direct correlation between output volume and percieved loudness. There appears to be a positive correlation though, of increasing distance between percieved volume against actual output volume. Nothing is detected before 5. Between 10 and 61 every note is at least one volume different to the previous note, though between 10 and 49 there are a seemingly random difference of one to 4 between volumes. After that, 50 to 61 have between 1 and 6 between volumes.

I have not repeated the test on these or any other computers. 10 to 61 seem to me the safest part of the Scratch volume to work with, with a difference of 4 between loudnesses to allow for difference in volume. (that's 25 characters per tone length.)

Please comment and try this at home!

Last edited by sparks (2011-01-06 15:56:05)