Speaker test "features" request

Speaker test "features" request

[Tom Watson]

> Date: Wed, 22 Jun 2005 18:22:17 +0800
> From: Raymond <rayau@netvigator.com>

<<<deletia>>>
> 
> I notice that there are female voice wav files in samples subdirectory.
> 
> ls speaker-test/samples
> Front_Center.wav  Makefile     Noise.wav        Rear_Right.wav 
> Side_Right.wav
> Front_Left.wav    Makefile.am  Rear_Center.wav  sample_map.csv
> Front_Right.wav   Makefile.in  Rear_Left.wav    Side_Left.wav
> 
> 
> Is there an option to use female voice instead of pink noise, sine wave ?
> 

As noted in the next message (removed) this isn't supported yet.  In doing
many "more than 2 channel" tests I offer the following suggestions and
observations  for implementation of a voice channel identification:
1)  Everyone calls the channels by different names
2)  Many people don't speak English.
3)  If you are poking around waiting for the identifiers to come back to
    the channel you are working on is a $%^&*%$#@ pain.
4)  Just call the channels "one" "two" "three", etc...
5)  Put out an id map in text when the voice identifier starts that
    indicates the correspondence.
6)  Voice ALL the channels at the same time.  You could be feeding a mixer
    board (or other similar instrument) and desire to get everything setup.
    Most mixer boards have a "solo" feature which is helpful in this regard.
7)  (MCP Comment) Female voice is nice.  Takes the geek edge off of it all.
8)  This is a really nice utility to have around.  It proves out the
    playback system.
9)  Perhaps we can get a "professional voice" to do this.  Fame and notoriety
    for the "Open source" voice.  Anyone have any contacts?


p.s.  None of this has anything to do with the error/bug report.  These are
just general comments about a "future direction".

-- 
Tom Watson
tsw@johana.com

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click

Re: Speaker test "features" request

[Takashi Iwai]

At Wed, 22 Jun 2005 04:26:17 -0700 (PDT),
Tom Watson wrote:
> 
> > Date: Wed, 22 Jun 2005 18:22:17 +0800
> > From: Raymond <rayau@netvigator.com>
> 
> <<<deletia>>>
> > 
> > I notice that there are female voice wav files in samples subdirectory.
> > 
> > ls speaker-test/samples
> > Front_Center.wav  Makefile     Noise.wav        Rear_Right.wav 
> > Side_Right.wav
> > Front_Left.wav    Makefile.am  Rear_Center.wav  sample_map.csv
> > Front_Right.wav   Makefile.in  Rear_Left.wav    Side_Left.wav
> > 
> > 
> > Is there an option to use female voice instead of pink noise, sine wave ?
> > 
> 
> As noted in the next message (removed) this isn't supported yet.  In doing
> many "more than 2 channel" tests I offer the following suggestions and
> observations  for implementation of a voice channel identification:
> 1)  Everyone calls the channels by different names
> 2)  Many people don't speak English.

The texts are already i18nized on CVS.  Provide an appropriate po file
for your locale.  Also, we can store voice files in different
directories per locale (if any).

> 3)  If you are poking around waiting for the identifiers to come back to
>     the channel you are working on is a $%^&*%$#@ pain.

Will be solved when (6) is implemeneted?

> 4)  Just call the channels "one" "two" "three", etc...

Doable, but I don't think we have much of this demand.

> 5)  Put out an id map in text when the voice identifier starts that
>     indicates the correspondence.

I think it's already so?

> 6)  Voice ALL the channels at the same time.  You could be feeding a mixer
>     board (or other similar instrument) and desire to get everything setup.
>     Most mixer boards have a "solo" feature which is helpful in this regard.

This would be nice as an option.

> 7)  (MCP Comment) Female voice is nice.  Takes the geek edge off of it all.

:)

> 8)  This is a really nice utility to have around.  It proves out the
>     playback system.
> 9)  Perhaps we can get a "professional voice" to do this.  Fame and notoriety
>     for the "Open source" voice.  Anyone have any contacts?

These are welcome, of course :)


Takashi


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click

Re: Speaker test "features" request

[Dirk Jagdmann]

> 4)  Just call the channels "one" "two" "three", etc...

I vote for such an option. I don't have any (standardized) x.1 speaker 
configuration and assign all my channels as needed and refer to them by 
numbers anyway.

> 6)  Voice ALL the channels at the same time.  You could be feeding a mixer

Does help debugging your cooked up wires/patchboard a lot.

> 9)  Perhaps we can get a "professional voice" to do this.  Fame and notoriety
>     for the "Open source" voice.  Anyone have any contacts?

The asterisk PhoneBox software includes samples in english from 1-20 and 
then the 10'er until 90, 100's, 1000's, 1M. So a (clever) software can 
construct numbers up to 10^9-1. As a start one could use them. Localized 
samples of the digits and numbers can then be added later on.

-- 
---> doj / cubic
----> http://cubic.org/~doj
-----> http://llg.cubic.org


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click