In relation to pronounciation corrections in eSpeak NG once more
Dejan Ristic
Hi to the respective community and users,
Due to the messages relating to the pronounciation corrections I happen to have read so far, I've just decided to extract them from the directory called Speech Dictionaries. They are contained in a notepad document, which implies that I may offer them whenever you like. The only question is where to post the document so that it may be accessed by the NVDA users. I do not consider myself to be supposed to do it here, for I have not obtained a permission to do so. I also do not think that it should be done that way because the mail traffic may be made lower. That's why I'm addressing myself to you so that I may do it in a proper manner so as not to come in breach of the rules and regulations of the mailing list. As I've already said, I've made over a hundred corrections. Let me confirm here that I've made 139 corrections so far. That's how users can help each other without calling a programmer to intervene. I hope that it is possible to find room for us the users to upload our corrections so that when accessing that piece of the site, we can automatically exchange pronounciation corrections. That's one way of how eSpeak NG speech may be updated without wating any next release to launch. If I'm supposed to send the mentioned document to an admin in a private message so that it may be reviewed, I'm glad to obey the rule. If I'm supposed to mention this topic no longer, I shall obey that rule, too. Your sincere friend and user, Dejan
|
|
Gene
I would think the best thing to do as far as making
the file available, is to copy the .dic file to another location on your
computer. If it's the default.dic file, that may be it's name. I've
almost never done this and others may be able to tell you where it is.
Then you could send it somewhere such as to a Drop Box folder or to someone who
will place it on their web site, such as the other Gene.
Gene
----- Original Message -----
From: Dejan Ristic
Sent: Friday, March 30, 2018 3:18 PM
Subject: [nvda] In relation to pronounciation corrections in eSpeak
NG once more Due to the messages relating to the pronounciation corrections I happen to have read so far, I've just decided to extract them from the directory called Speech Dictionaries. They are contained in a notepad document, which implies that I may offer them whenever you like. The only question is where to post the document so that it may be accessed by the NVDA users. I do not consider myself to be supposed to do it here, for I have not obtained a permission to do so. I also do not think that it should be done that way because the mail traffic may be made lower. That's why I'm addressing myself to you so that I may do it in a proper manner so as not to come in breach of the rules and regulations of the mailing list. As I've already said, I've made over a hundred corrections. Let me confirm here that I've made 139 corrections so far. That's how users can help each other without calling a programmer to intervene. I hope that it is possible to find room for us the users to upload our corrections so that when accessing that piece of the site, we can automatically exchange pronounciation corrections. That's one way of how eSpeak NG speech may be updated without wating any next release to launch. If I'm supposed to send the mentioned document to an admin in a private message so that it may be reviewed, I'm glad to obey the rule. If I'm supposed to mention this topic no longer, I shall obey that rule, too. Your sincere friend and user, Dejan
|
|
Sharni-Lee Ward
This sounds like a good idea, to be honest. Most of my dictionary entries are related to either Pokémon, names of characters from a couple other anime, and corrections to help Espeak handle compound words (to be fair, a lot of these are not usual compound words, but sometimes there were non-compound words rendered as such as well as the other way around). Some of these are personal to me, and someone not in the circle of Fanfiction.net that I was in would not have occasion to use them. There are names I've tried to get as close to accurate as possible, and sadly failed and aborted. I do not know how to naturally stress a syllable via regular expressions, so cannot get "Gallade" pronounced correctly, as an example.
That said, most people using NVDA have different synth preferences, as demonstrated by previous threads on this topic, so the synth being used as default should be noted when compiling the lists of Dictionary entries. But this would be a great resource to
put together.
On 31/03/2018 8:38 AM, Gene wrote:
|
|
Antony Stone
A wiki strikes me as being the ideal place to put such things.
toggle quoted messageShow quoted text
This would enable multiple contributors to create their own pages, with their own notes about which synth they use, what language / accent / dialect they prefer, and upload their version/s of dictionaries for others to try out and also learn from. I have no idea whether NVaccess has such a wiki as part of its online web resources, or would even want to as part of anything official, but if not, I'm sure there are people here on the list who could provide one if that's the best solution. Regards, Antony.
On Saturday 31 March 2018 at 01:12:09, Sharni-Lee Ward wrote:
This sounds like a good idea, to be honest. Most of my dictionary entries --
Angela Merkel arrives at Paris airport. "Nationality?" asks the immigration officer. "German," she replies. "Occupation?" "No, just here for a summit conference." Please reply to the list; please *don't* CC me.
|
|
Brian's Mail list account
Incidentally the one that still says newslitter rather than letter was fixed in Speaking, yet it seems not to be in the current nvda compiled version, so sometimes it seems that nvda is behind the releases of the synth.
toggle quoted messageShow quoted text
In the end the issue of doing it in dic files is a thorny one, for two reasons. 1. a user needs to know where to copy the file and also has to remember that any rules they have already defined will be lost.2. When a new version of Espeak does come out, you will not know how many of the changes have been implemented in the new version. There are many other pitfalls, not the least of which is crewing it up by shoving it in the wrong folder! One think I do think nvda needs is some kind of way to save and retrieve dic files for voices. Brian bglists@... Sent via blueyonder. Please address personal E-mail to:- briang1@..., putting 'Brian Gaff' in the display name field.
----- Original Message -----
From: "Gene" <gsasner@...> To: <nvda@nvda.groups.io> Sent: Friday, March 30, 2018 10:38 PM Subject: Re: [nvda] In relation to pronounciation corrections in eSpeak NG once more I would think the best thing to do as far as making the file available, is to copy the .dic file to another location on your computer. If it's the default.dic file, that may be it's name. I've almost never done this and others may be able to tell you where it is. Then you could send it somewhere such as to a Drop Box folder or to someone who will place it on their web site, such as the other Gene. Gene ----- Original Message ----- From: Dejan Ristic Sent: Friday, March 30, 2018 3:18 PM To: nvda@nvda.groups.io Subject: [nvda] In relation to pronounciation corrections in eSpeak NG once more Hi to the respective community and users, Due to the messages relating to the pronounciation corrections I happen to have read so far, I've just decided to extract them from the directory called Speech Dictionaries. They are contained in a notepad document, which implies that I may offer them whenever you like. The only question is where to post the document so that it may be accessed by the NVDA users. I do not consider myself to be supposed to do it here, for I have not obtained a permission to do so. I also do not think that it should be done that way because the mail traffic may be made lower. That's why I'm addressing myself to you so that I may do it in a proper manner so as not to come in breach of the rules and regulations of the mailing list. As I've already said, I've made over a hundred corrections. Let me confirm here that I've made 139 corrections so far. That's how users can help each other without calling a programmer to intervene. I hope that it is possible to find room for us the users to upload our corrections so that when accessing that piece of the site, we can automatically exchange pronounciation corrections. That's one way of how eSpeak NG speech may be updated without wating any next release to launch. If I'm supposed to send the mentioned document to an admin in a private message so that it may be reviewed, I'm glad to obey the rule. If I'm supposed to mention this topic no longer, I shall obey that rule, too. Your sincere friend and user, Dejan
|
|