- ... I understand foldback and aliasing, but don t quite see why it s much of an issue with speech or even music when sampling at 44-Ksps, as there isMessage 1 of 126 , Dec 5, 2010View Source--- In SeattleRobotics@yahoogroups.com, "Peter Balch" <peterbalch@...> wrote:
>I understand foldback and aliasing, but don't quite see why
> > Another difference is that I wanted to be able to display data
> > at as high a fequency as I could, whereas with something like
> > speech+music 98% [whatever] of the energy is well below the
> > maximum sampleable nyquist frequency.
> True but those high freqs are very important. Look at the difference between > Kar and Tar or Sar and SHar. Most therapy seems to be with children and they > have significant energy above 10kHz. Therapists often concentrate on those > kinds of phonemes.
> 44100 sounds generous but turns out not to be. We reckon a sharp cutoff at > 20kHz is just about tolerable. (As I'm sure you know, if we don't filter > before the ADC then the very high frequecies of speech "fold back" over the > spectrum and mess it up.)
it's much of an issue with speech or even music when sampling at
44-Ksps, as there is essentially no energy near or above 20-Hz
for those signals. I suppose you can tell the difference by
>The holding capacitor values come straight out of the Microchip
> > > TACQ which is (according to which part of the doc you read)
> > > either 100nS, 7.67uS or 11uS.
> What do you think it actually is?
> > I assume the "F" series
> > went to 120-pF from 51-pF on the "C" series because they wanted a
> > little better noise filtering for 10-bit res,
> Interesting. where did you get those figures? There must be a doc I haven't > read.
datasheets. For the PIC16F87X, d/s DS39582B, the cap is shown
in figure 11-2, and TACQ is also on that page. The "C" series
cap value also came out of the d/s.
I notice some of d/s for later parts don't show these values.
Looking at the formula, the TACQ value is obviously based upon
a 1st-order capacitor charging ckt, and is linearly related to
both the value CHOLD and also the impedance of the external
voltage source, RS.
TACQ = Amplifier Settling Time + Hold Capacitor Charging Time
+ Temperature Coefficient
= TAMP + TC + TCOFF
TC = 2 ìs + TC + [(Temperature 25°C)(0.05 ìs/°C)]
= CHOLD (RIC + RSS + RS) In(1/2047)
= - 120 pF (1 kÙ + 7 kÙ + 10 kÙ) In(0.0004885)
= 16.47 ìs
TACQ = 2 ìs + 16.47 ìs + [(50°C 25°C)(0.05 ìs/°C)
= 19.72 ìs
TACQ is dependent directly upon RS. The other R values in the
formula are internal components. If RS is close to zero, then
TACQ will be about 10-usec, so it varies dependent upon the
external ckt. CHOLD is 51-pF in the "C" series, so TACQ will be
less than 1/2 of the value shown here.
- You mean the SPO256 by GI, 28pin DIL ? They also had a 40 pin DIL CTS256A-AL2 text to speech for spo256 I have use the SPO256 back in the 80s and also lastMessage 126 of 126 , Dec 12, 2010View SourceYou mean the SPO256 by GI, 28pin DIL ?They also had a 40 pin DIL CTS256A-AL2 text to speech for spo256I have use the SPO256 back in the 80s and also last month but never got around to using the latter, I do have the Tandy data that came with them (somewhere I hope).David - robots42----- Original Message -----From: Rich ChandlerSent: Sunday, December 12, 2010 11:26 PMSubject: Re: [SeattleRobotics] Re: Speech and frequency vs time
A long time ago, when Radio Shack still sold IC's, I got a speech
synthesizer set, but I never did anything with it. Has anyone used them?