Richard Owlett wrote:
> Rocky wrote:
>
>> Jon Harris wrote:
>>
>>> "Jerry Avins" <[email protected]> wrote in message
>>> news:[email protected]..
>>>
>>>> Richard Owlett wrote:
>>>>
>>>>> I think I know the answers, but then again
>>>>>
>>>>> I have a "hobby" project to evaluate intelligibility of speech with
>>>>> varying
>>>>> parameters ( primarily sample rate and bandwidth ).
>>>>>
>>>>> [ for perspective, in a previous incantation I "discovered"
>>>>> formants ;]
>>>>>
>>>>> My sample universe is a reading of the Bible recorded under studio
>>>>> conditions. As it is a commercial recording, I'm assuming Nyquist
>>>>> is the
>>>>> least restrictive criterion it meets.
>>>>>
>>>>> Am I correct thinking that if I digitally filter my source to less
>>>>> than 3 kHz
>>>>> i can safely decimate to 8 kHz sample rate?
>>>>
>>>>
>>>> Yes
>>>
>>>
>>> And if you filter is good enough, you can push the cut-off further up
>>> toward
>>> 4kHz.
>>
>>
>>
>> You might want to consider a slightly heretical approach and set the
>> filter ABOVE nyquist! Use aliasing as one of the parameters. Try about
>> 5KHz cuttoff with 8KHz sampling rate.
>>
>> I think you might be surprised at the results if it is only speech you
>> are evaluating.
>>
>> Regards
>> Rocky
>>
>
> Moi. Heretical?
> All I've done is suggest DSP be done in TECO, COBOL, or VBA in MSWord ;}

You missed one of the critically important ones - Excel and OpenOffice
spreadsheets. :-)

In the Lotus 123 days I did a lot of DSP experimentation in
spreadsheets. I had some fairly usable filter design stuff too. I wonder
if that is still on an old disk somwhere?

Steve