moderated Vocalizer and difference in pronunciation


Van Lant, Robin
 

Hi,

I’m using Vocalizer Nathan Compact with JAWS 2021.  This is not the first case, but is striking today, so it prompted me to pose the question.  I was reading through an Excel file for work of incoming new hires.  I thought there was a typo, because the name of one of the colleges was mispronounced if I read the cell.  If I read word by word, it says the word fine.  I pasted the name into Word and the same thing occurs. Reading in any manner except word by word, causes Vocalizer to mispronounce it.  I’ve switched to Eloquence, Microsoft Mobile, and VoiceWare, all of which do not have this issue.  As I noted, this is not the first case.  Do other people notice similar issues with Vocalizer?  Is there a technical reason why a word would be pronounced differently in word by word versus line by line or say all? 

 

Here is the college.  The last word is the issue.  I have fixed it in my dictionary manager for now by adding a third s to the word.

College of the Holy Cross

 

I really love the sound  of Nathan’s compact voice, but am quite surprised by some of the strange pronunciations it has for common words.  Even the opposite of “Female” is totally off and is the first thing I adjust in a new dictionary file.

 

Using Windows 10, JAWS 2021, Office 365.

Robin Van Lant

 




KeyCorp Public This communication may contain privileged and/or confidential information. It is intended solely for the use of the addressee. No confidentiality or privilege is waived or lost as a result of any transmission errors. If you are not the intended recipient, you are strictly prohibited from disclosing, copying, distributing or using any of this information. If you have received this communication in error, please contact the sender immediately and destroy the material in its entirety, whether electronic or hard copy.

This communication is for informational purposes only, is not an offer, solicitation, recommendation or commitment for any transaction or to buy or sell any security or other financial product, and is not intended as investment advice or as a confirmation of any transaction. Any market price, indicative value, estimate, view, opinion, data or other information herein is not warranted as to completeness or accuracy, is subject to change without notice, and KeyBanc Capital Markets Inc. accepts no liability for its use or to update or keep it current. Any views or opinions are those of the individual sender, not necessarily of KeyBanc Capital Markets Inc.

The sender of this communication is a licensed securities representative employed by or associated with KeyBanc Capital Markets Inc. Member FINRA/SIPC and may also represent KeyBank National Association (“KeyBank N.A.”). Securities products and services are offered by KeyBanc Capital Markets Inc. Banking products and services are offered by KeyBank N.A.

127 Public Square, Cleveland, OH 44114

If you prefer not to receive future e-mail offers for products or services from Key send an e-mail to DNERequests@... with 'No Promotional E-mails' in the SUBJECT line.


Soronel Haetir
 

Yes, I have noticed this. Sometimes the synthesizer is even correct
(English does have words that are pronounced differently depending on
context, though this does not strike me as such a case.

Eloquence was made to be usable at very high speeds while sacrificing
'naturalness', Vocalizer is intended to sound natural but cannot
achieve nearly the speed that Eloquence can and in the end it
generally does a 'good enough' job.

And yes, Samantha also mispronounces it in this instance, saying
something like 'croass'. There are a few '--ross' words that have an
'oh' but doing a quick search through a words list they are greatly
outnumbered by those using an 'aw'.

On 2/1/21, Van Lant, Robin via groups.io
<Robin_Van_Lant=Key.com@groups.io> wrote:
Hi,
I'm using Vocalizer Nathan Compact with JAWS 2021. This is not the first
case, but is striking today, so it prompted me to pose the question. I was
reading through an Excel file for work of incoming new hires. I thought
there was a typo, because the name of one of the colleges was mispronounced
if I read the cell. If I read word by word, it says the word fine. I
pasted the name into Word and the same thing occurs. Reading in any manner
except word by word, causes Vocalizer to mispronounce it. I've switched to
Eloquence, Microsoft Mobile, and VoiceWare, all of which do not have this
issue. As I noted, this is not the first case. Do other people notice
similar issues with Vocalizer? Is there a technical reason why a word would
be pronounced differently in word by word versus line by line or say all?

Here is the college. The last word is the issue. I have fixed it in my
dictionary manager for now by adding a third s to the word.
College of the Holy Cross

I really love the sound of Nathan's compact voice, but am quite surprised
by some of the strange pronunciations it has for common words. Even the
opposite of "Female" is totally off and is the first thing I adjust in a new
dictionary file.

Using Windows 10, JAWS 2021, Office 365.
Robin Van Lant




KeyCorp Public This communication may contain privileged and/or confidential
information. It is intended solely for the use of the addressee. No
confidentiality or privilege is waived or lost as a result of any
transmission errors. If you are not the intended recipient, you are strictly
prohibited from disclosing, copying, distributing or using any of this
information. If you have received this communication in error, please
contact the sender immediately and destroy the material in its entirety,
whether electronic or hard copy.

This communication is for informational purposes only, is not an offer,
solicitation, recommendation or commitment for any transaction or to buy or
sell any security or other financial product, and is not intended as
investment advice or as a confirmation of any transaction. Any market price,
indicative value, estimate, view, opinion, data or other information herein
is not warranted as to completeness or accuracy, is subject to change
without notice, and KeyBanc Capital Markets Inc. accepts no liability for
its use or to update or keep it current. Any views or opinions are those of
the individual sender, not necessarily of KeyBanc Capital Markets Inc.

The sender of this communication is a licensed securities representative
employed by or associated with KeyBanc Capital Markets Inc. Member
FINRA/SIPC and may also represent KeyBank National Association ("KeyBank
N.A."). Securities products and services are offered by KeyBanc Capital
Markets Inc. Banking products and services are offered by KeyBank N.A.

127 Public Square, Cleveland, OH 44114

If you prefer not to receive future e-mail offers for products or services
from Key send an e-mail to DNERequests@key.com with 'No Promotional E-mails'
in the SUBJECT line.






--
Soronel Haetir
soronel.haetir@gmail.com


Chris Hill
 

Hello.


Yes, synthesizers try to guess context, if they just pronounced words, they wouldn't work near as well.  It becomes obvious when they guess wrong.  May I ask why you are using the compact version of the voice?  These are always inferior to the real versions, and most hardware won't have trouble with the full one.



On 2/1/2021 17:22, Van Lant, Robin via groups.io wrote:

Hi,

I’m using Vocalizer Nathan Compact with JAWS 2021.  This is not the first case, but is striking today, so it prompted me to pose the question.  I was reading through an Excel file for work of incoming new hires.  I thought there was a typo, because the name of one of the colleges was mispronounced if I read the cell.  If I read word by word, it says the word fine.  I pasted the name into Word and the same thing occurs. Reading in any manner except word by word, causes Vocalizer to mispronounce it.  I’ve switched to Eloquence, Microsoft Mobile, and VoiceWare, all of which do not have this issue.  As I noted, this is not the first case.  Do other people notice similar issues with Vocalizer?  Is there a technical reason why a word would be pronounced differently in word by word versus line by line or say all? 

 

Here is the college.  The last word is the issue.  I have fixed it in my dictionary manager for now by adding a third s to the word.

College of the Holy Cross

 

I really love the sound  of Nathan’s compact voice, but am quite surprised by some of the strange pronunciations it has for common words.  Even the opposite of “Female” is totally off and is the first thing I adjust in a new dictionary file.

 

Using Windows 10, JAWS 2021, Office 365.

Robin Van Lant

 




KeyCorp Public This communication may contain privileged and/or confidential information. It is intended solely for the use of the addressee. No confidentiality or privilege is waived or lost as a result of any transmission errors. If you are not the intended recipient, you are strictly prohibited from disclosing, copying, distributing or using any of this information. If you have received this communication in error, please contact the sender immediately and destroy the material in its entirety, whether electronic or hard copy.

This communication is for informational purposes only, is not an offer, solicitation, recommendation or commitment for any transaction or to buy or sell any security or other financial product, and is not intended as investment advice or as a confirmation of any transaction. Any market price, indicative value, estimate, view, opinion, data or other information herein is not warranted as to completeness or accuracy, is subject to change without notice, and KeyBanc Capital Markets Inc. accepts no liability for its use or to update or keep it current. Any views or opinions are those of the individual sender, not necessarily of KeyBanc Capital Markets Inc.

The sender of this communication is a licensed securities representative employed by or associated with KeyBanc Capital Markets Inc. Member FINRA/SIPC and may also represent KeyBank National Association (“KeyBank N.A.”). Securities products and services are offered by KeyBanc Capital Markets Inc. Banking products and services are offered by KeyBank N.A.

127 Public Square, Cleveland, OH 44114

If you prefer not to receive future e-mail offers for products or services from Key send an e-mail to DNERequests@... with 'No Promotional E-mails' in the SUBJECT line.


Soronel Haetir
 

In this case even the full versions have problems, at least Samantha does.

On 2/1/21, Chris Hill <hillco@earthlink.net> wrote:
Hello.


Yes, synthesizers try to guess context, if they just pronounced words,
they wouldn't work near as well.  It becomes obvious when they guess
wrong.  May I ask why you are using the compact version of the voice?
These are always inferior to the real versions, and most hardware won't
have trouble with the full one.



On 2/1/2021 17:22, Van Lant, Robin via groups.io wrote:

Hi,

I’m using Vocalizer Nathan Compact with JAWS 2021.  This is not the
first case, but is striking today, so it prompted me to pose the
question.  I was reading through an Excel file for work of incoming
new hires.  I thought there was a typo, because the name of one of the
colleges was mispronounced if I read the cell.  If I read word by
word, it says the word fine.  I pasted the name into Word and the same
thing occurs. Reading in any manner except word by word, causes
Vocalizer to mispronounce it.  I’ve switched to Eloquence, Microsoft
Mobile, and VoiceWare, all of which do not have this issue.  As I
noted, this is not the first case. Do other people notice similar
issues with Vocalizer?  Is there a technical reason why a word would
be pronounced differently in word by word versus line by line or say all?

Here is the college.  The last word is the issue.  I have fixed it in
my dictionary manager for now by adding a third s to the word.

College of the Holy Cross

I really love the sound  of Nathan’s compact voice, but am quite
surprised by some of the strange pronunciations it has for common
words.  Even the opposite of “Female” is totally off and is the first
thing I adjust in a new dictionary file.

Using Windows 10, JAWS 2021, Office 365.

*Robin Van Lant*




KeyCorp Public This communication may contain privileged and/or
confidential information. It is intended solely for the use of the
addressee. No confidentiality or privilege is waived or lost as a
result of any transmission errors. If you are not the intended
recipient, you are strictly prohibited from disclosing, copying,
distributing or using any of this information. If you have received
this communication in error, please contact the sender immediately and
destroy the material in its entirety, whether electronic or hard copy.

This communication is for informational purposes only, is not an
offer, solicitation, recommendation or commitment for any transaction
or to buy or sell any security or other financial product, and is not
intended as investment advice or as a confirmation of any transaction.
Any market price, indicative value, estimate, view, opinion, data or
other information herein is not warranted as to completeness or
accuracy, is subject to change without notice, and KeyBanc Capital
Markets Inc. accepts no liability for its use or to update or keep it
current. Any views or opinions are those of the individual sender, not
necessarily of KeyBanc Capital Markets Inc.

The sender of this communication is a licensed securities
representative employed by or associated with KeyBanc Capital Markets
Inc. Member FINRA/SIPC and may also represent KeyBank National
Association (“KeyBank N.A.”). Securities products and services are
offered by KeyBanc Capital Markets Inc. Banking products and services
are offered by KeyBank N.A.

127 Public Square, Cleveland, OH 44114

If you prefer not to receive future e-mail offers for products or
services from Key send an e-mail to DNERequests@key.com with 'No
Promotional E-mails' in the SUBJECT line.




--
Soronel Haetir
soronel.haetir@gmail.com


Van Lant, Robin
 

Chris,

It’s been a while, but, even on my iPhone, I think there is more inflection in the full or enhanced versions.  I was a Window Eyes user with Dec Talk for many years, so I don’t mind the robotic sounds of Eloquence, but there is something to the about the voices that I find hard or fuzzy to my hearing, not the diction, but the audio quality.  I like the sound clarity in Vocalizer Nathan the most, despite some of its pronunciation quirks. I like that the Compact voices are less emotive and more steady.  I’m also not one who listens to speech super-fast, so that isn’t a concern to me. I’m faster than the default for sure, but not the rocket speed I hear others use. I’m not diagnosed as hearing impaired, but have some sort of issues with my hearing of certain frequencies or tones.  I also like Microsoft Mobile Mike for clarity.

 

 

From: main@jfw.groups.io <main@jfw.groups.io> On Behalf Of Chris Hill
Sent: Monday, February 1, 2021 5:46 PM
To: main@jfw.groups.io
Subject: Re: Vocalizer and difference in pronunciation

 

WARNING: This email originated externally. Exercise caution. Think before clicking links or opening attachments.

 

Hello.

 

Yes, synthesizers try to guess context, if they just pronounced words, they wouldn't work near as well.  It becomes obvious when they guess wrong.  May I ask why you are using the compact version of the voice?  These are always inferior to the real versions, and most hardware won't have trouble with the full one.

 

 

On 2/1/2021 17:22, Van Lant, Robin via groups.io wrote:

Hi,

I’m using Vocalizer Nathan Compact with JAWS 2021.  This is not the first case, but is striking today, so it prompted me to pose the question.  I was reading through an Excel file for work of incoming new hires.  I thought there was a typo, because the name of one of the colleges was mispronounced if I read the cell.  If I read word by word, it says the word fine.  I pasted the name into Word and the same thing occurs. Reading in any manner except word by word, causes Vocalizer to mispronounce it.  I’ve switched to Eloquence, Microsoft Mobile, and VoiceWare, all of which do not have this issue.  As I noted, this is not the first case.  Do other people notice similar issues with Vocalizer?  Is there a technical reason why a word would be pronounced differently in word by word versus line by line or say all? 

 

Here is the college.  The last word is the issue.  I have fixed it in my dictionary manager for now by adding a third s to the word.

College of the Holy Cross

 

I really love the sound  of Nathan’s compact voice, but am quite surprised by some of the strange pronunciations it has for common words.  Even the opposite of “Female” is totally off and is the first thing I adjust in a new dictionary file.

 

Using Windows 10, JAWS 2021, Office 365.

Robin Van Lant

 




KeyCorp Public This communication may contain privileged and/or confidential information. It is intended solely for the use of the addressee. No confidentiality or privilege is waived or lost as a result of any transmission errors. If you are not the intended recipient, you are strictly prohibited from disclosing, copying, distributing or using any of this information. If you have received this communication in error, please contact the sender immediately and destroy the material in its entirety, whether electronic or hard copy.

This communication is for informational purposes only, is not an offer, solicitation, recommendation or commitment for any transaction or to buy or sell any security or other financial product, and is not intended as investment advice or as a confirmation of any transaction. Any market price, indicative value, estimate, view, opinion, data or other information herein is not warranted as to completeness or accuracy, is subject to change without notice, and KeyBanc Capital Markets Inc. accepts no liability for its use or to update or keep it current. Any views or opinions are those of the individual sender, not necessarily of KeyBanc Capital Markets Inc.

The sender of this communication is a licensed securities representative employed by or associated with KeyBanc Capital Markets Inc. Member FINRA/SIPC and may also represent KeyBank National Association (“KeyBank N.A.”). Securities products and services are offered by KeyBanc Capital Markets Inc. Banking products and services are offered by KeyBank N.A.

127 Public Square, Cleveland, OH 44114

If you prefer not to receive future e-mail offers for products or services from Key send an e-mail to DNERequests@... with 'No Promotional E-mails' in the SUBJECT line.


Chris Hill
 

Robin:


I don't like eloquence either.  The version that used to ship with k1000 was better, but the one with jaws sounds like audio above a certain frequency is completely missing.  I use Tom Premium High on Jaws and my phone both, it seems about the best, and since I can have it on both, things are usually mis-pronounced everywhere, which bothers me less.


I used to have a Dec express, I wish I could get that voice back, maybe without the problem of not being able to shut it up quickly, that might be the best of all worlds.



On 2/2/2021 10:07, Van Lant, Robin via groups.io wrote:

Chris,

It’s been a while, but, even on my iPhone, I think there is more inflection in the full or enhanced versions.  I was a Window Eyes user with Dec Talk for many years, so I don’t mind the robotic sounds of Eloquence, but there is something to the about the voices that I find hard or fuzzy to my hearing, not the diction, but the audio quality.  I like the sound clarity in Vocalizer Nathan the most, despite some of its pronunciation quirks. I like that the Compact voices are less emotive and more steady.  I’m also not one who listens to speech super-fast, so that isn’t a concern to me. I’m faster than the default for sure, but not the rocket speed I hear others use. I’m not diagnosed as hearing impaired, but have some sort of issues with my hearing of certain frequencies or tones.  I also like Microsoft Mobile Mike for clarity.

 

 

From: main@jfw.groups.io <main@jfw.groups.io> On Behalf Of Chris Hill
Sent: Monday, February 1, 2021 5:46 PM
To: main@jfw.groups.io
Subject: Re: Vocalizer and difference in pronunciation

 

WARNING: This email originated externally. Exercise caution. Think before clicking links or opening attachments.

 

Hello.

 

Yes, synthesizers try to guess context, if they just pronounced words, they wouldn't work near as well.  It becomes obvious when they guess wrong.  May I ask why you are using the compact version of the voice?  These are always inferior to the real versions, and most hardware won't have trouble with the full one.

 

 

On 2/1/2021 17:22, Van Lant, Robin via groups.io wrote:

Hi,

I’m using Vocalizer Nathan Compact with JAWS 2021.  This is not the first case, but is striking today, so it prompted me to pose the question.  I was reading through an Excel file for work of incoming new hires.  I thought there was a typo, because the name of one of the colleges was mispronounced if I read the cell.  If I read word by word, it says the word fine.  I pasted the name into Word and the same thing occurs. Reading in any manner except word by word, causes Vocalizer to mispronounce it.  I’ve switched to Eloquence, Microsoft Mobile, and VoiceWare, all of which do not have this issue.  As I noted, this is not the first case.  Do other people notice similar issues with Vocalizer?  Is there a technical reason why a word would be pronounced differently in word by word versus line by line or say all? 

 

Here is the college.  The last word is the issue.  I have fixed it in my dictionary manager for now by adding a third s to the word.

College of the Holy Cross

 

I really love the sound  of Nathan’s compact voice, but am quite surprised by some of the strange pronunciations it has for common words.  Even the opposite of “Female” is totally off and is the first thing I adjust in a new dictionary file.

 

Using Windows 10, JAWS 2021, Office 365.

Robin Van Lant

 




KeyCorp Public This communication may contain privileged and/or confidential information. It is intended solely for the use of the addressee. No confidentiality or privilege is waived or lost as a result of any transmission errors. If you are not the intended recipient, you are strictly prohibited from disclosing, copying, distributing or using any of this information. If you have received this communication in error, please contact the sender immediately and destroy the material in its entirety, whether electronic or hard copy.

This communication is for informational purposes only, is not an offer, solicitation, recommendation or commitment for any transaction or to buy or sell any security or other financial product, and is not intended as investment advice or as a confirmation of any transaction. Any market price, indicative value, estimate, view, opinion, data or other information herein is not warranted as to completeness or accuracy, is subject to change without notice, and KeyBanc Capital Markets Inc. accepts no liability for its use or to update or keep it current. Any views or opinions are those of the individual sender, not necessarily of KeyBanc Capital Markets Inc.

The sender of this communication is a licensed securities representative employed by or associated with KeyBanc Capital Markets Inc. Member FINRA/SIPC and may also represent KeyBank National Association (“KeyBank N.A.”). Securities products and services are offered by KeyBanc Capital Markets Inc. Banking products and services are offered by KeyBank N.A.

127 Public Square, Cleveland, OH 44114

If you prefer not to receive future e-mail offers for products or services from Key send an e-mail to DNERequests@... with 'No Promotional E-mails' in the SUBJECT line.


JM Casey
 

I thought K1000 shipped with software DecTalk? At least, that’s what it sounds like. I had it in the early 2000s and it was sure not Eloquence then.

 

I like Eloquence for every day use. The reason you can’t hear those high frequencies is – well, they’re not there. It’s an 11khz synth. It’s been around forever and comes from a time when that was pretty n ormal, and there hasn’t really been any need to change it. I have the IBM TTS for nVDA that sounds a lot like Eloquence, but withs ome small pronunciation differences and a little bit more frequency response.

 

 

 

From: main@jfw.groups.io <main@jfw.groups.io> On Behalf Of Chris Hill
Sent: February 2, 2021 01:08 PM
To: main@jfw.groups.io
Subject: Re: Vocalizer and difference in pronunciation

 

Robin:

 

I don't like eloquence either.  The version that used to ship with k1000 was better, but the one with jaws sounds like audio above a certain frequency is completely missing.  I use Tom Premium High on Jaws and my phone both, it seems about the best, and since I can have it on both, things are usually mis-pronounced everywhere, which bothers me less.

 

I used to have a Dec express, I wish I could get that voice back, maybe without the problem of not being able to shut it up quickly, that might be the best of all worlds.

 

 

On 2/2/2021 10:07, Van Lant, Robin via groups.io wrote:

Chris,

It’s been a while, but, even on my iPhone, I think there is more inflection in the full or enhanced versions.  I was a Window Eyes user with Dec Talk for many years, so I don’t mind the robotic sounds of Eloquence, but there is something to the about the voices that I find hard or fuzzy to my hearing, not the diction, but the audio quality.  I like the sound clarity in Vocalizer Nathan the most, despite some of its pronunciation quirks. I like that the Compact voices are less emotive and more steady.  I’m also not one who listens to speech super-fast, so that isn’t a concern to me. I’m faster than the default for sure, but not the rocket speed I hear others use. I’m not diagnosed as hearing impaired, but have some sort of issues with my hearing of certain frequencies or tones.  I also like Microsoft Mobile Mike for clarity.

 

 

From: main@jfw.groups.io <main@jfw.groups.io> On Behalf Of Chris Hill
Sent: Monday, February 1, 2021 5:46 PM
To: main@jfw.groups.io
Subject: Re: Vocalizer and difference in pronunciation

 

WARNING: This email originated externally. Exercise caution. Think before clicking links or opening attachments.

 

Hello.

 

Yes, synthesizers try to guess context, if they just pronounced words, they wouldn't work near as well.  It becomes obvious when they guess wrong.  May I ask why you are using the compact version of the voice?  These are always inferior to the real versions, and most hardware won't have trouble with the full one.

 

 

On 2/1/2021 17:22, Van Lant, Robin via groups.io wrote:

Hi,

I’m using Vocalizer Nathan Compact with JAWS 2021.  This is not the first case, but is striking today, so it prompted me to pose the question.  I was reading through an Excel file for work of incoming new hires.  I thought there was a typo, because the name of one of the colleges was mispronounced if I read the cell.  If I read word by word, it says the word fine.  I pasted the name into Word and the same thing occurs. Reading in any manner except word by word, causes Vocalizer to mispronounce it.  I’ve switched to Eloquence, Microsoft Mobile, and VoiceWare, all of which do not have this issue.  As I noted, this is not the first case.  Do other people notice similar issues with Vocalizer?  Is there a technical reason why a word would be pronounced differently in word by word versus line by line or say all? 

 

Here is the college.  The last word is the issue.  I have fixed it in my dictionary manager for now by adding a third s to the word.

College of the Holy Cross

 

I really love the sound  of Nathan’s compact voice, but am quite surprised by some of the strange pronunciations it has for common words.  Even the opposite of “Female” is totally off and is the first thing I adjust in a new dictionary file.

 

Using Windows 10, JAWS 2021, Office 365.

Robin Van Lant

 




KeyCorp Public This communication may contain privileged and/or confidential information. It is intended solely for the use of the addressee. No confidentiality or privilege is waived or lost as a result of any transmission errors. If you are not the intended recipient, you are strictly prohibited from disclosing, copying, distributing or using any of this information. If you have received this communication in error, please contact the sender immediately and destroy the material in its entirety, whether electronic or hard copy.

This communication is for informational purposes only, is not an offer, solicitation, recommendation or commitment for any transaction or to buy or sell any security or other financial product, and is not intended as investment advice or as a confirmation of any transaction. Any market price, indicative value, estimate, view, opinion, data or other information herein is not warranted as to completeness or accuracy, is subject to change without notice, and KeyBanc Capital Markets Inc. accepts no liability for its use or to update or keep it current. Any views or opinions are those of the individual sender, not necessarily of KeyBanc Capital Markets Inc.

The sender of this communication is a licensed securities representative employed by or associated with KeyBanc Capital Markets Inc. Member FINRA/SIPC and may also represent KeyBank National Association (“KeyBank N.A.”). Securities products and services are offered by KeyBanc Capital Markets Inc. Banking products and services are offered by KeyBank N.A.

127 Public Square, Cleveland, OH 44114

If you prefer not to receive future e-mail offers for products or services from Key send an e-mail to DNERequests@... with 'No Promotional E-mails' in the SUBJECT line.


David Griffith
 

No Kurzweil nowadays comes with ETI Eloquence and Voice Ware Kate Paul and latterly James in later versions (AKA Neospeech).

I started out with version 11 I think but am on version 14 now. It is possible Dec Talk was available in earlier incarnations.

I have never had access to Dec Talk with Kurzweil but would have been interested in giving it a try if it was available.

David G.

 

Sent from Mail for Windows 10

 

From: JM Casey
Sent: 02 February 2021 18:45
To: main@jfw.groups.io
Subject: Re: Vocalizer and difference in pronunciation

 

I thought K1000 shipped with software DecTalk? At least, that’s what it sounds like. I had it in the early 2000s and it was sure not Eloquence then.

 

I like Eloquence for every day use. The reason you can’t hear those high frequencies is – well, they’re not there. It’s an 11khz synth. It’s been around forever and comes from a time when that was pretty n ormal, and there hasn’t really been any need to change it. I have the IBM TTS for nVDA that sounds a lot like Eloquence, but withs ome small pronunciation differences and a little bit more frequency response.

 

 

 

From: main@jfw.groups.io <main@jfw.groups.io> On Behalf Of Chris Hill
Sent: February 2, 2021 01:08 PM
To: main@jfw.groups.io
Subject: Re: Vocalizer and difference in pronunciation

 

Robin:

 

I don't like eloquence either.  The version that used to ship with k1000 was better, but the one with jaws sounds like audio above a certain frequency is completely missing.  I use Tom Premium High on Jaws and my phone both, it seems about the best, and since I can have it on both, things are usually mis-pronounced everywhere, which bothers me less.

 

I used to have a Dec express, I wish I could get that voice back, maybe without the problem of not being able to shut it up quickly, that might be the best of all worlds.

 

 

On 2/2/2021 10:07, Van Lant, Robin via groups.io wrote:

Chris,

It’s been a while, but, even on my iPhone, I think there is more inflection in the full or enhanced versions.  I was a Window Eyes user with Dec Talk for many years, so I don’t mind the robotic sounds of Eloquence, but there is something to the about the voices that I find hard or fuzzy to my hearing, not the diction, but the audio quality.  I like the sound clarity in Vocalizer Nathan the most, despite some of its pronunciation quirks. I like that the Compact voices are less emotive and more steady.  I’m also not one who listens to speech super-fast, so that isn’t a concern to me. I’m faster than the default for sure, but not the rocket speed I hear others use. I’m not diagnosed as hearing impaired, but have some sort of issues with my hearing of certain frequencies or tones.  I also like Microsoft Mobile Mike for clarity.

 

 

From: main@jfw.groups.io <main@jfw.groups.io> On Behalf Of Chris Hill
Sent: Monday, February 1, 2021 5:46 PM
To: main@jfw.groups.io
Subject: Re: Vocalizer and difference in pronunciation

 

WARNING: This email originated externally. Exercise caution. Think before clicking links or opening attachments.

 

Hello.

 

Yes, synthesizers try to guess context, if they just pronounced words, they wouldn't work near as well.  It becomes obvious when they guess wrong.  May I ask why you are using the compact version of the voice?  These are always inferior to the real versions, and most hardware won't have trouble with the full one.

 

 

On 2/1/2021 17:22, Van Lant, Robin via groups.io wrote:

Hi,

I’m using Vocalizer Nathan Compact with JAWS 2021.  This is not the first case, but is striking today, so it prompted me to pose the question.  I was reading through an Excel file for work of incoming new hires.  I thought there was a typo, because the name of one of the colleges was mispronounced if I read the cell.  If I read word by word, it says the word fine.  I pasted the name into Word and the same thing occurs. Reading in any manner except word by word, causes Vocalizer to mispronounce it.  I’ve switched to Eloquence, Microsoft Mobile, and VoiceWare, all of which do not have this issue.  As I noted, this is not the first case.  Do other people notice similar issues with Vocalizer?  Is there a technical reason why a word would be pronounced differently in word by word versus line by line or say all? 

 

Here is the college.  The last word is the issue.  I have fixed it in my dictionary manager for now by adding a third s to the word.

College of the Holy Cross

 

I really love the sound  of Nathan’s compact voice, but am quite surprised by some of the strange pronunciations it has for common words.  Even the opposite of “Female” is totally off and is the first thing I adjust in a new dictionary file.

 

Using Windows 10, JAWS 2021, Office 365.

Robin Van Lant

 




KeyCorp Public This communication may contain privileged and/or confidential information. It is intended solely for the use of the addressee. No confidentiality or privilege is waived or lost as a result of any transmission errors. If you are not the intended recipient, you are strictly prohibited from disclosing, copying, distributing or using any of this information. If you have received this communication in error, please contact the sender immediately and destroy the material in its entirety, whether electronic or hard copy.

This communication is for informational purposes only, is not an offer, solicitation, recommendation or commitment for any transaction or to buy or sell any security or other financial product, and is not intended as investment advice or as a confirmation of any transaction. Any market price, indicative value, estimate, view, opinion, data or other information herein is not warranted as to completeness or accuracy, is subject to change without notice, and KeyBanc Capital Markets Inc. accepts no liability for its use or to update or keep it current. Any views or opinions are those of the individual sender, not necessarily of KeyBanc Capital Markets Inc.

The sender of this communication is a licensed securities representative employed by or associated with KeyBanc Capital Markets Inc. Member FINRA/SIPC and may also represent KeyBank National Association (“KeyBank N.A.”). Securities products and services are offered by KeyBanc Capital Markets Inc. Banking products and services are offered by KeyBank N.A.

127 Public Square, Cleveland, OH 44114

If you prefer not to receive future e-mail offers for products or services from Key send an e-mail to DNERequests@... with 'No Promotional E-mails' in the SUBJECT line.

 


JM Casey
 

Oh yeah, this was a much older version, and it ran on my Windows 98 machine. It was the exact same synth as could be found in the old reading machiens that Kurzweil made, which I’m sure was Dectalk. Too bad they gave it up as that was a rather nice synth, especially for something that came about in the 1980s and basically never changed, audio-wise.

 

 

From: main@jfw.groups.io <main@jfw.groups.io> On Behalf Of David Griffith
Sent: February 2, 2021 01:54 PM
To: main@jfw.groups.io
Subject: Re: Vocalizer and difference in pronunciation

 

No Kurzweil nowadays comes with ETI Eloquence and Voice Ware Kate Paul and latterly James in later versions (AKA Neospeech).

I started out with version 11 I think but am on version 14 now. It is possible Dec Talk was available in earlier incarnations.

I have never had access to Dec Talk with Kurzweil but would have been interested in giving it a try if it was available.

David G.

 

Sent from Mail for Windows 10

 

From: JM Casey
Sent: 02 February 2021 18:45
To: main@jfw.groups.io
Subject: Re: Vocalizer and difference in pronunciation

 

I thought K1000 shipped with software DecTalk? At least, that’s what it sounds like. I had it in the early 2000s and it was sure not Eloquence then.

 

I like Eloquence for every day use. The reason you can’t hear those high frequencies is – well, they’re not there. It’s an 11khz synth. It’s been around forever and comes from a time when that was pretty n ormal, and there hasn’t really been any need to change it. I have the IBM TTS for nVDA that sounds a lot like Eloquence, but withs ome small pronunciation differences and a little bit more frequency response.

 

 

 

From: main@jfw.groups.io <main@jfw.groups.io> On Behalf Of Chris Hill
Sent: February 2, 2021 01:08 PM
To: main@jfw.groups.io
Subject: Re: Vocalizer and difference in pronunciation

 

Robin:

 

I don't like eloquence either.  The version that used to ship with k1000 was better, but the one with jaws sounds like audio above a certain frequency is completely missing.  I use Tom Premium High on Jaws and my phone both, it seems about the best, and since I can have it on both, things are usually mis-pronounced everywhere, which bothers me less.

 

I used to have a Dec express, I wish I could get that voice back, maybe without the problem of not being able to shut it up quickly, that might be the best of all worlds.

 

 

On 2/2/2021 10:07, Van Lant, Robin via groups.io wrote:

Chris,

It’s been a while, but, even on my iPhone, I think there is more inflection in the full or enhanced versions.  I was a Window Eyes user with Dec Talk for many years, so I don’t mind the robotic sounds of Eloquence, but there is something to the about the voices that I find hard or fuzzy to my hearing, not the diction, but the audio quality.  I like the sound clarity in Vocalizer Nathan the most, despite some of its pronunciation quirks. I like that the Compact voices are less emotive and more steady.  I’m also not one who listens to speech super-fast, so that isn’t a concern to me. I’m faster than the default for sure, but not the rocket speed I hear others use. I’m not diagnosed as hearing impaired, but have some sort of issues with my hearing of certain frequencies or tones.  I also like Microsoft Mobile Mike for clarity.

 

 

From: main@jfw.groups.io <main@jfw.groups.io> On Behalf Of Chris Hill
Sent: Monday, February 1, 2021 5:46 PM
To: main@jfw.groups.io
Subject: Re: Vocalizer and difference in pronunciation

 

WARNING: This email originated externally. Exercise caution. Think before clicking links or opening attachments.

 

Hello.

 

Yes, synthesizers try to guess context, if they just pronounced words, they wouldn't work near as well.  It becomes obvious when they guess wrong.  May I ask why you are using the compact version of the voice?  These are always inferior to the real versions, and most hardware won't have trouble with the full one.

 

 

On 2/1/2021 17:22, Van Lant, Robin via groups.io wrote:

Hi,

I’m using Vocalizer Nathan Compact with JAWS 2021.  This is not the first case, but is striking today, so it prompted me to pose the question.  I was reading through an Excel file for work of incoming new hires.  I thought there was a typo, because the name of one of the colleges was mispronounced if I read the cell.  If I read word by word, it says the word fine.  I pasted the name into Word and the same thing occurs. Reading in any manner except word by word, causes Vocalizer to mispronounce it.  I’ve switched to Eloquence, Microsoft Mobile, and VoiceWare, all of which do not have this issue.  As I noted, this is not the first case.  Do other people notice similar issues with Vocalizer?  Is there a technical reason why a word would be pronounced differently in word by word versus line by line or say all? 

 

Here is the college.  The last word is the issue.  I have fixed it in my dictionary manager for now by adding a third s to the word.

College of the Holy Cross

 

I really love the sound  of Nathan’s compact voice, but am quite surprised by some of the strange pronunciations it has for common words.  Even the opposite of “Female” is totally off and is the first thing I adjust in a new dictionary file.

 

Using Windows 10, JAWS 2021, Office 365.

Robin Van Lant

 




KeyCorp Public This communication may contain privileged and/or confidential information. It is intended solely for the use of the addressee. No confidentiality or privilege is waived or lost as a result of any transmission errors. If you are not the intended recipient, you are strictly prohibited from disclosing, copying, distributing or using any of this information. If you have received this communication in error, please contact the sender immediately and destroy the material in its entirety, whether electronic or hard copy.

This communication is for informational purposes only, is not an offer, solicitation, recommendation or commitment for any transaction or to buy or sell any security or other financial product, and is not intended as investment advice or as a confirmation of any transaction. Any market price, indicative value, estimate, view, opinion, data or other information herein is not warranted as to completeness or accuracy, is subject to change without notice, and KeyBanc Capital Markets Inc. accepts no liability for its use or to update or keep it current. Any views or opinions are those of the individual sender, not necessarily of KeyBanc Capital Markets Inc.

The sender of this communication is a licensed securities representative employed by or associated with KeyBanc Capital Markets Inc. Member FINRA/SIPC and may also represent KeyBank National Association (“KeyBank N.A.”). Securities products and services are offered by KeyBanc Capital Markets Inc. Banking products and services are offered by KeyBank N.A.

127 Public Square, Cleveland, OH 44114

If you prefer not to receive future e-mail offers for products or services from Key send an e-mail to DNERequests@... with 'No Promotional E-mails' in the SUBJECT line.

 


Van Lant, Robin
 

Glad to understand the frequency aspect of why I might have issues hearing Eloquence clearly.  Ever so often I’ll test out voices again, but most have either too much inflection or must have some of this frequency difference.  Even in Vocalizer, I find the U.S. voices other than Nathan and Samantha seem more muffled to me.  Vocalizer.  

 

Robin

 

 

From: main@jfw.groups.io <main@jfw.groups.io> On Behalf Of Chris Hill
Sent: Tuesday, February 2, 2021 11:08 AM
To: main@jfw.groups.io
Subject: Re: Vocalizer and difference in pronunciation

 

Robin:

 

I don't like eloquence either.  The version that used to ship with k1000 was better, but the one with jaws sounds like audio above a certain frequency is completely missing.  I use Tom Premium High on Jaws and my phone both, it seems about the best, and since I can have it on both, things are usually mis-pronounced everywhere, which bothers me less.

 

I used to have a Dec express, I wish I could get that voice back, maybe without the problem of not being able to shut it up quickly, that might be the best of all worlds.

 

 

On 2/2/2021 10:07, Van Lant, Robin via groups.io wrote:

Chris,

It’s been a while, but, even on my iPhone, I think there is more inflection in the full or enhanced versions.  I was a Window Eyes user with Dec Talk for many years, so I don’t mind the robotic sounds of Eloquence, but there is something to the about the voices that I find hard or fuzzy to my hearing, not the diction, but the audio quality.  I like the sound clarity in Vocalizer Nathan the most, despite some of its pronunciation quirks. I like that the Compact voices are less emotive and more steady.  I’m also not one who listens to speech super-fast, so that isn’t a concern to me. I’m faster than the default for sure, but not the rocket speed I hear others use. I’m not diagnosed as hearing impaired, but have some sort of issues with my hearing of certain frequencies or tones.  I also like Microsoft Mobile Mike for clarity.

 

 

From: main@jfw.groups.io <main@jfw.groups.io> On Behalf Of Chris Hill
Sent: Monday, February 1, 2021 5:46 PM
To: main@jfw.groups.io
Subject: Re: Vocalizer and difference in pronunciation

 

WARNING: This email originated externally. Exercise caution. Think before clicking links or opening attachments.

 

Hello.

 

Yes, synthesizers try to guess context, if they just pronounced words, they wouldn't work near as well.  It becomes obvious when they guess wrong.  May I ask why you are using the compact version of the voice?  These are always inferior to the real versions, and most hardware won't have trouble with the full one.

 

 

On 2/1/2021 17:22, Van Lant, Robin via groups.io wrote:

Hi,

I’m using Vocalizer Nathan Compact with JAWS 2021.  This is not the first case, but is striking today, so it prompted me to pose the question.  I was reading through an Excel file for work of incoming new hires.  I thought there was a typo, because the name of one of the colleges was mispronounced if I read the cell.  If I read word by word, it says the word fine.  I pasted the name into Word and the same thing occurs. Reading in any manner except word by word, causes Vocalizer to mispronounce it.  I’ve switched to Eloquence, Microsoft Mobile, and VoiceWare, all of which do not have this issue.  As I noted, this is not the first case.  Do other people notice similar issues with Vocalizer?  Is there a technical reason why a word would be pronounced differently in word by word versus line by line or say all? 

 

Here is the college.  The last word is the issue.  I have fixed it in my dictionary manager for now by adding a third s to the word.

College of the Holy Cross

 

I really love the sound  of Nathan’s compact voice, but am quite surprised by some of the strange pronunciations it has for common words.  Even the opposite of “Female” is totally off and is the first thing I adjust in a new dictionary file.

 

Using Windows 10, JAWS 2021, Office 365.

Robin Van Lant

 




KeyCorp Public This communication may contain privileged and/or confidential information. It is intended solely for the use of the addressee. No confidentiality or privilege is waived or lost as a result of any transmission errors. If you are not the intended recipient, you are strictly prohibited from disclosing, copying, distributing or using any of this information. If you have received this communication in error, please contact the sender immediately and destroy the material in its entirety, whether electronic or hard copy.

This communication is for informational purposes only, is not an offer, solicitation, recommendation or commitment for any transaction or to buy or sell any security or other financial product, and is not intended as investment advice or as a confirmation of any transaction. Any market price, indicative value, estimate, view, opinion, data or other information herein is not warranted as to completeness or accuracy, is subject to change without notice, and KeyBanc Capital Markets Inc. accepts no liability for its use or to update or keep it current. Any views or opinions are those of the individual sender, not necessarily of KeyBanc Capital Markets Inc.

The sender of this communication is a licensed securities representative employed by or associated with KeyBanc Capital Markets Inc. Member FINRA/SIPC and may also represent KeyBank National Association (“KeyBank N.A.”). Securities products and services are offered by KeyBanc Capital Markets Inc. Banking products and services are offered by KeyBank N.A.

127 Public Square, Cleveland, OH 44114

If you prefer not to receive future e-mail offers for products or services from Key send an e-mail to DNERequests@... with 'No Promotional E-mails' in the SUBJECT line.


JM Casey
 

I do understand. I have a hearing loss myself and when I’m watching movies or listening to audiodramas, especially old ones, I tend to use equaliser settings to bring out more of the high frequencies in dialogue. As for the Eloquence synth, it’s top “resolution’ if you want to call it that is 11khz roughly, and the rule of thumb is that you get half that in terms of top-most frequency audible, so around 5.5khz, which is still *(supposed* to be above the frequency of most connsonants, but in practical terms, your ear can probably still discern that something is missing.

I still think Eloquence is the best synth for everyday computer use because it’s more concerned with smoothness and practical use than actual pleasurable reading. I don’t set it to speak very fast, myself, but still much faster than Vocalizer, whihch, as it’s taken from actual human voice samples, sounds like absolute garbage speaking any faster than a person can speak. However I do switch to vocaliser when I want to listen to a book “on the go” or something with my bluetooth headset. Otherwise though I just stick to refreshable braille most of the time.

 

From: main@jfw.groups.io <main@jfw.groups.io> On Behalf Of Van Lant, Robin via groups.io
Sent: February 2, 2021 04:33 PM
To: main@jfw.groups.io
Subject: Re: Vocalizer and difference in pronunciation

 

Glad to understand the frequency aspect of why I might have issues hearing Eloquence clearly.  Ever so often I’ll test out voices again, but most have either too much inflection or must have some of this frequency difference.  Even in Vocalizer, I find the U.S. voices other than Nathan and Samantha seem more muffled to me.  Vocalizer.  

 

Robin

 

 

From: main@jfw.groups.io <main@jfw.groups.io> On Behalf Of Chris Hill
Sent: Tuesday, February 2, 2021 11:08 AM
To: main@jfw.groups.io
Subject: Re: Vocalizer and difference in pronunciation

 

Robin:

 

I don't like eloquence either.  The version that used to ship with k1000 was better, but the one with jaws sounds like audio above a certain frequency is completely missing.  I use Tom Premium High on Jaws and my phone both, it seems about the best, and since I can have it on both, things are usually mis-pronounced everywhere, which bothers me less.

 

I used to have a Dec express, I wish I could get that voice back, maybe without the problem of not being able to shut it up quickly, that might be the best of all worlds.

 

 

On 2/2/2021 10:07, Van Lant, Robin via groups.io wrote:

Chris,

It’s been a while, but, even on my iPhone, I think there is more inflection in the full or enhanced versions.  I was a Window Eyes user with Dec Talk for many years, so I don’t mind the robotic sounds of Eloquence, but there is something to the about the voices that I find hard or fuzzy to my hearing, not the diction, but the audio quality.  I like the sound clarity in Vocalizer Nathan the most, despite some of its pronunciation quirks. I like that the Compact voices are less emotive and more steady.  I’m also not one who listens to speech super-fast, so that isn’t a concern to me. I’m faster than the default for sure, but not the rocket speed I hear others use. I’m not diagnosed as hearing impaired, but have some sort of issues with my hearing of certain frequencies or tones.  I also like Microsoft Mobile Mike for clarity.

 

 

From: main@jfw.groups.io <main@jfw.groups.io> On Behalf Of Chris Hill
Sent: Monday, February 1, 2021 5:46 PM
To: main@jfw.groups.io
Subject: Re: Vocalizer and difference in pronunciation

 

WARNING: This email originated externally. Exercise caution. Think before clicking links or opening attachments.

 

Hello.

 

Yes, synthesizers try to guess context, if they just pronounced words, they wouldn't work near as well.  It becomes obvious when they guess wrong.  May I ask why you are using the compact version of the voice?  These are always inferior to the real versions, and most hardware won't have trouble with the full one.

 

 

On 2/1/2021 17:22, Van Lant, Robin via groups.io wrote:

Hi,

I’m using Vocalizer Nathan Compact with JAWS 2021.  This is not the first case, but is striking today, so it prompted me to pose the question.  I was reading through an Excel file for work of incoming new hires.  I thought there was a typo, because the name of one of the colleges was mispronounced if I read the cell.  If I read word by word, it says the word fine.  I pasted the name into Word and the same thing occurs. Reading in any manner except word by word, causes Vocalizer to mispronounce it.  I’ve switched to Eloquence, Microsoft Mobile, and VoiceWare, all of which do not have this issue.  As I noted, this is not the first case.  Do other people notice similar issues with Vocalizer?  Is there a technical reason why a word would be pronounced differently in word by word versus line by line or say all? 

 

Here is the college.  The last word is the issue.  I have fixed it in my dictionary manager for now by adding a third s to the word.

College of the Holy Cross

 

I really love the sound  of Nathan’s compact voice, but am quite surprised by some of the strange pronunciations it has for common words.  Even the opposite of “Female” is totally off and is the first thing I adjust in a new dictionary file.

 

Using Windows 10, JAWS 2021, Office 365.

Robin Van Lant

 




KeyCorp Public This communication may contain privileged and/or confidential information. It is intended solely for the use of the addressee. No confidentiality or privilege is waived or lost as a result of any transmission errors. If you are not the intended recipient, you are strictly prohibited from disclosing, copying, distributing or using any of this information. If you have received this communication in error, please contact the sender immediately and destroy the material in its entirety, whether electronic or hard copy.

This communication is for informational purposes only, is not an offer, solicitation, recommendation or commitment for any transaction or to buy or sell any security or other financial product, and is not intended as investment advice or as a confirmation of any transaction. Any market price, indicative value, estimate, view, opinion, data or other information herein is not warranted as to completeness or accuracy, is subject to change without notice, and KeyBanc Capital Markets Inc. accepts no liability for its use or to update or keep it current. Any views or opinions are those of the individual sender, not necessarily of KeyBanc Capital Markets Inc.

The sender of this communication is a licensed securities representative employed by or associated with KeyBanc Capital Markets Inc. Member FINRA/SIPC and may also represent KeyBank National Association (“KeyBank N.A.”). Securities products and services are offered by KeyBanc Capital Markets Inc. Banking products and services are offered by KeyBank N.A.

127 Public Square, Cleveland, OH 44114

If you prefer not to receive future e-mail offers for products or services from Key send an e-mail to DNERequests@... with 'No Promotional E-mails' in the SUBJECT line.