Replies: 5 comments 3 replies
-
Voice names are case sensitive. vitaliy Is Vitaliy.
From: Denis ***@***.***>
Sent: Monday, October 2, 2023 12:05 PM
To: RHVoice/RHVoice ***@***.***>
Cc: Subscribed ***@***.***>
Subject: [RHVoice/RHVoice] How to make use of rhvoice.conf? (Discussion #796)
Okay, at first I misunderstood and thought that the conf file should affect RHVoice-test (#795 <#795> ). But now I have configured speech-dispatcher, but spd-say still uses the wrong voice.
My RHVoice.conf:
quality=max
default_volume=1.5
voices.vitaliy.default_pitch=1.0
voices.bdl.default_pitch=1.2
voices.vitaliy.default_rate=0.85
voices.bdl.default_rate=0.8
voice_profiles=vitaliy+bdl
When I run spd-say -L it does list all the profiles available for RHVoice (imcluding Vitaliy+Bdl, as specified in the conf file):
NAME LANGUAGE VARIANT
Alan en none
Aleksandr-hq ru none
Anna ru none
Artemiy ru none
Bdl en none
Clb en none
Evgeniy-Eng en none
Evgeniy-Rus ru none
Lyubov en none
Mikhail ru none
Pavel ru none
Slt en none
Vitaliy ru none
Vitaliy+Bdl ru none
Yuriy ru none
But if I run spd-say "some text", it still says the text in the wrong voice. E.g. if I run spd-say "Проверка русской речи", the voice used is Aleksandr-hq, with default pitch, rate, and volume.
My speechd.conf: https://dpaste.org/SjhMa
—
Reply to this email directly, view it on GitHub <#796> , or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACVCDE74RFPN7V4N5Z5NJPLX5KGWBANCNFSM6AAAAAA5PFWUQQ> .
You are receiving this because you are subscribed to this thread. <https://github.com/notifications/beacon/ACVCDE55ONW6C5LRUOAZLFLX5KGWBA5CNFSM6AAAAAA5PFWUQSWGG33NNVSW45C7OR4XAZNKIRUXGY3VONZWS33OVJRW63LNMVXHIX3JMTHAAVW6FY.gif> Message ID: ***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
You need to switch voice.
From: Denis ***@***.***>
Sent: Monday, October 2, 2023 12:37 PM
To: RHVoice/RHVoice ***@***.***>
Cc: Zvonimir Stanečić ***@***.***>; Comment ***@***.***>
Subject: Re: [RHVoice/RHVoice] How to make use of rhvoice.conf? (Discussion #796)
I have modified the conf file so that voices start with the capital letters, but it didn't help. spd-say "some text" still uses Aleksandr-hq for Russian and Alan for English.
—
Reply to this email directly, view it on GitHub <#796 (reply in thread)> , or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACVCDE7OA3IKVAK4LMZBQ73X5KKNTANCNFSM6AAAAAA5PFWUQQ> .
You are receiving this because you commented. <https://github.com/notifications/beacon/ACVCDE2XERE3LMQRO4I7EYTX5KKNTA5CNFSM6AAAAAA5PFWUQSWGG33NNVSW45C7OR4XAZNRIRUXGY3VONZWS33OINXW23LFNZ2KUY3PNVWWK3TUL5UWJTQANVHOA.gif> Message ID: ***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
You need to switch the voice, because you are setting up voice profiles in the configuration file.
You are specifying these for the use by the speech dispatcher or any other client that walks and talks on earth with RHVoice.
Default rate is the rate specified by the incremental values of 50 percent. So, rate should be specified, too.
Specifying deult rate means specifying rate of 50 percent
From: Denis ***@***.***>
Sent: Monday, October 2, 2023 1:04 PM
To: RHVoice/RHVoice ***@***.***>
Cc: Zvonimir Stanečić ***@***.***>; Comment ***@***.***>
Subject: Re: [RHVoice/RHVoice] How to make use of rhvoice.conf? (Discussion #796)
You mean use the spd-say -y somevoice? But then why have a config file at all? Also, even if I use the needed voices via this switch, their pitch and rate are still default (1.0), and not those specified in the config file.
—
Reply to this email directly, view it on GitHub <#796 (reply in thread)> , or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACVCDE62IWR3ZCWMLQG7EELX5KNR7ANCNFSM6AAAAAA5PFWUQQ> .
You are receiving this because you commented. <https://github.com/notifications/beacon/ACVCDE7B3TSUVVL5OBWCH3LX5KNR7A5CNFSM6AAAAAA5PFWUQSWGG33NNVSW45C7OR4XAZNRIRUXGY3VONZWS33OINXW23LFNZ2KUY3PNVWWK3TUL5UWJTQANVH32.gif> Message ID: ***@***.*** ***@***.***> >
|
Beta Was this translation helpful? Give feedback.
-
No, it is used, but in different way you expect.
From: Denis ***@***.***>
Sent: Monday, October 2, 2023 1:31 PM
To: RHVoice/RHVoice ***@***.***>
Cc: Zvonimir Stanečić ***@***.***>; Comment ***@***.***>
Subject: Re: [RHVoice/RHVoice] How to make use of rhvoice.conf? (Discussion #796)
So, what, RHVoice.conf is unused? Since you have to specify everything with command-line switches anyway?
—
Reply to this email directly, view it on GitHub <#796 (reply in thread)> , or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACVCDE5JGZ6NCJCZ4BSYXLLX5KQW3ANCNFSM6AAAAAA5PFWUQQ> .
You are receiving this because you commented. <https://github.com/notifications/beacon/ACVCDEZNMMJHBEJD6IU6EL3X5KQW3A5CNFSM6AAAAAA5PFWUQSWGG33NNVSW45C7OR4XAZNRIRUXGY3VONZWS33OINXW23LFNZ2KUY3PNVWWK3TUL5UWJTQANVIIS.gif> Message ID: ***@***.*** ***@***.***> >
|
Beta Was this translation helpful? Give feedback.
-
Okay, maybe it's better to phrase my question from the desired result perspective. I have a text mostly in Russian but with insterspersed English phrases. Splitting it into separate Russian and English parts would be ineffective because English sections are very short --- like I said, only phrases and even separate words. So I need to process the whole text with 2 voices: a Russian voice and an English voice. However, I would also prefer to adjust the pitch and rate for these voices separately: for English voice (Bdl) 1.2 and 0.8, and for Russian voice (Vitaliy) 1.0 and 0.85. Doing so by command-line switches for RHVoice-test is impossible because it applies the same pitch and rate specified by -t and -r switches to both voices; and like I said, I need different settings per voice. I thought I could use RHVoice.conf for this purpose because in there you can specify settings per voice, but that isn't working. So, is it possible to transfer text to audio, using different voices for Russian and English and appying different pitch and rate for these voices? |
Beta Was this translation helpful? Give feedback.
-
Okay, at first I misunderstood and thought that the conf file should affect RHVoice-test (#795). But now I have configured speech-dispatcher, but spd-say still uses the wrong voice.
My RHVoice.conf:
When I run
spd-say -L
it does list all the profiles available for RHVoice (imcluding Vitaliy+Bdl, as specified in the conf file):But if I run
spd-say "some text"
, it still says the text in the wrong voice. E.g. if I runspd-say "Проверка русской речи"
, the voice used is Aleksandr-hq, with default pitch, rate, and volume.My speechd.conf: https://dpaste.org/SjhMa
Beta Was this translation helpful? Give feedback.
All reactions