1 |
James <wireless <at> tampabay.rr.com> writes: |
2 |
|
3 |
|
4 |
> lee <lee <at> yagibdah.de> writes: |
5 |
|
6 |
You know, it just dawned on me a solution that *may* fit your needs. |
7 |
|
8 |
If folks where to first employ a speech to text interpolate software, |
9 |
on their (originating) end, then your friend would receive mostly accurate |
10 |
text communications. The single source (voice) solution becomes more accurate |
11 |
over time as one uses them. If this sort of semantic occurred before the |
12 |
text was sent to your friend, then it's much easier to support and |
13 |
refine. |
14 |
|
15 |
|
16 |
Sadly, it does nothing for new sources of contact, until they refine their |
17 |
own speech-to-text system. |
18 |
|
19 |
Sometimes a simultaneous video feed help folks comprehend, by reading lips |
20 |
whilst discerning audio signals, but that requires low-latency connections. |
21 |
|
22 |
|
23 |
hth, |
24 |
James |