I got that feeling that spoken word was not an option while looking
through the literature for Capstan. Was hoping someone on the list had
experience enough to know and I'm glad for your response before I spent
more time playing around with that program.
I'll have to listen/review closely but it seemed to me there was no
"constant" tone in the background or some type of hum that the Plagent
Process could "key" into when I viewed it on a spectrum analyzer.
My guess is that the recorder had an older belt or motor that seemed to
"warm up" over time and helped drive the tape more steadily as time went
on into the recording. The cassette itself does not appear to have any
binding issues that would have originally caused this. I've tried three
different decks and also rehoused the cassette, to insure it wasn't
anything on my end. Two cassettes had this issue, the other cassettes
from the same collection were fine.
I'd have to play back the digital version again to closely examine the
timing of the wow, but I think it's pretty consistent. I might be able
to build a custom pitch curve in Wavelab for this but I didn't want to
go through the effort if there was something out there that was easy to
work with and would have sufficed. Not on a high-end mastering level but
on a listenable level that was intelligible.
I just mentioned the $250 as that was what I was willing to spend on
some type of software plugin that might make things sound a bit better
on the two tapes that have the issue. I was going to do that out of my
own pocket as a courtesy to the client as the content was important to
them and they're already over budget. It is a collection from a small
college in the Northwest US, and the content is interviews with Japanese
Americans that were interned during WWII. If you listen really closely,
you can still make out what people are saying most of the time on these
two troubled tapes but it's not something you cannot casually listen to,
you have to concentrate hard.
If anyone else has a solution, I'd love to hear it.
Media Transfer Service, LLC
On 8/26/2018 8:09 PM, Tim Gillett wrote:
> Hi John,
> Unfortunately it appears Capstan only works on "polyphonic" musical
> sources, not on speech. We had a discussion about this same issue on
> the Sound on Sound forum and the company's product manager came on,
> saying he had tried but been been unable to successfully reduce wow on
> speech. All the company's references to successful removal of wow seem
> to apply to music only.
> Jamie Howarth's "Plagent Process" uses the recorded bias tone as a
> speed reference. That normally only applies to professional open reel
> recordings but if there are other tones on the recording such as a
> whistle etc it may be possible to use that as a speed reference to at
> least partially fix the wow.
> Unusual to have wow at the beginning of a tape side. Usually it will
> be worse at the end of a tape side.
> ----- Original Message ----- From: "John Schroth"
> <[log in to unmask]>
> To: <[log in to unmask]>
> Sent: Sunday, August 26, 2018 11:43 PM
> Subject: [ARSCLIST] Audio Cassette WOW repair
>> Hello ARSC members. I'm hoping to get some advice.
>> I'm digitizing a collection of aural history interviews recorded on
>> cassette tape. The deck that was used for recording was a standard
>> low cost portable deck with the built-in mic. Several of the tapes
>> have some serious wow at the beginning of the tape. It gets better 5
>> minutes in although it's still there as the recording goes on.
>> I'm hoping to repair the wow with at least an acceptable listenable
>> result through software. I read a bit about Melodyne Capstan, but it
>> appears to me that this program keys in on musical notes to fix wow
>> in content that's musical in nature. Since this content is not
>> musical, would Capstan still work? The price for the software is
>> beyond the client's budget, so I'm happy to work with someone who has
>> the software, if the software would work and that person would be
>> interested in a small one-off job.
>> Are there other programs that can automatically fix wow in recorded
>> content that is spoken word? There's too much content time to try and
>> fix this manually and still stay within the client's budget
>> parameters so it has to be an automatic software based program that
>> is affordable ($250 or less). I can't seem to find other offerings
>> that fit within the client's budget parameters.
>> Thanks in advance for any input.
>> Kind Regards,
>> John Schroth
>> Media Transfer Service, LLC
> This email has been checked for viruses by Avast antivirus software.
> This email has been checked for viruses by AVG.
Media Transfer Service, LLC
High Quality Conversion Of:
Video - Audio - Motion Picture - Still Image
Find out what's new at MTS: