Date   

[espeak-ng:master] new issue: Support for timing output #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng:master] New Issue Created by cmrdt:
#173 Support for timing output

maybe options to output subtitle-subrip (.srt) format files with option to split lines by word or line or phoneme(consonant-vowels) maybe also IPA/phonetic output and syntax coloring/formatting(font,bold)

https://en.wikipedia.org/wiki/SubRip#SubRip_text_file_format https//en.wikipedia.org/wiki/SubStation_Alpha

also, espeak (not NG) ttsapp's synth-"mouth-position" GUI is synced with the speech


[espeak-ng:master] new issue: Parse into formatted input/output #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng:master] New Issue Created by cmrdt:
#174 Parse into formatted input/output

to help readability (custom ASCII phonetic multi-glyph symbols hard to read) color-coding syntax highlighting to assist readability &debugging

maybe making the RGB color a function

maybe https://en.wikipedia.org/wiki/YUV or https://en.wikipedia.org/wiki/Lab_color_space

maybe also support for IPA input with https://en.wikipedia.org/wiki/International_Phonetic_Alphabet#Diacritics_and_prosodic_notation (more/less round, voiced, central)

maybe interface GUI also similar to formant.tcl demo in (install active-tcl8.4 also) http://www.speech.kth.se/snack/


[espeak-ng:master] reported: GUI for real-time custom voice-variant parameter tuning #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng:master] New Comment on Issue #175 GUI for real-time custom voice-variant parameter tuning
By cmrdt:

maybe something similar to

clipboard

clipboard2


Re: Updated Wikipedia article about eSpeakNG

abdelwaheb turki <turkiabdelwaheb@...>
 

Dear Sir,

I thank you for your answer. I will use http://reecedunn.co.uk/espeak-ng/ in the English Wikipedia when it will be finally created.

Yours Sincerely,

Houcemeddine Turki


De : espeak-ng@groups.io <espeak-ng@groups.io> de la part de Reece H. Dunn <msclrhd@...>
Envoyé : mercredi 28 décembre 2016 17:50
À : espeak-ng@groups.io
Objet : Re: [espeak-ng] Updated Wikipedia article about eSpeakNG
 
On Wed, Dec 28, 2016 at 08:32 am, abdelwaheb turki wrote:

Dear Sir,

I thank you for your answer. I will review that. However, please create http://espeakng.sourceforge.net so that it can be used.

eSpeak-ng is not hosted on SourceForge, so it does not make sense to create an espeakng.sourceforge.net page. Currently, the best location for this would be https://github.com/espeak-ng. I have http://reecedunn.co.uk/espeak-for-android, but that is focused toward the Android port. I will create an http://reecedunn.co.uk/espeak-ng page later on (likely around the 7th Jan when I have access to my home computer), which would then be a better fit.


Kind regards,

Reece H. Dunn


Yours Sincerely,

Houcemeddine Turki


De : espeak-ng@groups.io <espeak-ng@groups.io> de la part de Valdis Vitolins <valdis.vitolins@...>
Envoyé : mercredi 28 décembre 2016 15:10
À : espeak-ng@groups.io
Objet : Re: [espeak-ng] Updated Wikipedia article about eSpeakNG
 
https://en.wikipedia.org/wiki/ESpeakNG#Synthesis_method sections including multimedia samples.

As English is not my native language, review will be very appreciated.

Valdis

 


Re: Updated Wikipedia article about eSpeakNG

Reece H. Dunn
 

On Wed, Dec 28, 2016 at 08:32 am, abdelwaheb turki wrote:

Dear Sir,

I thank you for your answer. I will review that. However, please create http://espeakng.sourceforge.net so that it can be used.

eSpeak-ng is not hosted on SourceForge, so it does not make sense to create an espeakng.sourceforge.net page. Currently, the best location for this would be https://github.com/espeak-ng. I have http://reecedunn.co.uk/espeak-for-android, but that is focused toward the Android port. I will create an http://reecedunn.co.uk/espeak-ng page later on (likely around the 7th Jan when I have access to my home computer), which would then be a better fit.


Kind regards,

Reece H. Dunn


Yours Sincerely,

Houcemeddine Turki


De : espeak-ng@groups.io <espeak-ng@groups.io> de la part de Valdis Vitolins <valdis.vitolins@...>
Envoyé : mercredi 28 décembre 2016 15:10
À : espeak-ng@groups.io
Objet : Re: [espeak-ng] Updated Wikipedia article about eSpeakNG
 
https://en.wikipedia.org/wiki/ESpeakNG#Synthesis_method sections including multimedia samples.

As English is not my native language, review will be very appreciated.

Valdis

 


Re: Updated Wikipedia article about eSpeakNG

abdelwaheb turki <turkiabdelwaheb@...>
 

Dear Sir,

I thank you for your answer. I will review that. However, please create http://espeakng.sourceforge.net so that it can be used.

Yours Sincerely,

Houcemeddine Turki


De : espeak-ng@groups.io <espeak-ng@groups.io> de la part de Valdis Vitolins <valdis.vitolins@...>
Envoyé : mercredi 28 décembre 2016 15:10
À : espeak-ng@groups.io
Objet : Re: [espeak-ng] Updated Wikipedia article about eSpeakNG
 
https://en.wikipedia.org/wiki/ESpeakNG#Synthesis_method sections including multimedia samples.

As English is not my native language, review will be very appreciated.

Valdis


Re: Updated Wikipedia article about eSpeakNG

Valdis Vitolins
 

https://en.wikipedia.org/wiki/ESpeakNG#Synthesis_method sections including multimedia samples.

As English is not my native language, review will be very appreciated.

Valdis


Updates to Github #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng:master] New Issue Created by cmrdt:
#179 multi-channel voice-component output for DSPs, mixing or debugging

add command option to split voice into multiple channels?

  • voice-tone
  • voice-tone-roughness
  • consonants voiced
  • consonants unvoiced
  • breath

(time/pitch stretching, noise removal, compression, remastering)


[espeak-ng:master] New Issue Created by cmrdt:
#175 GUI for real-time custom voice-variant parameter tuning

the wide array of numbers(e.g. voice variant files) are hard to all optimize by hand manually what about sliders and/or 2D canvas objects, - a deformed mesh to somehow represent the array of param's?, - maybe also opacity for formant strength - stacked-bar-graph for max&min frequencies - conditional formatting with both audio &visual feedback -real-time tuning of wpm


[espeak-ng:master] reported: upgrade internal audio quality (sample rate and bit-depth) #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng:master] New Comment on Issue #178 upgrade internal audio quality (sample rate and bit-depth)
By cmrdt:

also if audio file same quality as device native-sampling rate, it would not have to resample again

also android resamplers use up battery and may be of low quality.. detect native sampling rate maybe by playing a sine wave chirp at various sample-rates and looking at the spectrogram


Updates to Github #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng:master] New Issue Created by cmrdt:
#178 upgrade internal audio quality (sample rate and bit-depth)

also in espeakedit (not ng) the resampling quality seemed to be so-so (maybe not sox's fault, maybe command-line-settings) but resampling from a low-quality source-wav file isn't the same as starting from one of higher quality (lower rates lack higher frequencies)

default currently is 11025×2 Hz 16-bits maybe increase to (18×2)kHz+ and 32-bit-float as compile option; & rather do downsampling from higher quality if need be (update syllables containing high frequencies: s f z ʃ ʒ t k θ ð v d )

and if using sox use best-quality sinc option, max precision


[espeak-ng:master] New Comment on Issue #178 upgrade internal audio quality (sample rate and bit-depth)
By cmrdt:

also if audio file same quality as device native-sampling rate, it would not have to resample again

also android resamplers use up battery and may be of low quality..


[espeak-ng:master] New Comment on Issue #173 Support for timing output
By cmrdt:

one of possible uses - if for media players (e.g. VLC) which support loading these subtitles (auto if same filename) and have it sync to text to speech


[espeak-ng:master] New Issue Created by cmrdt:
#179 multi-channel voice output for DSPs, mixing or debugging

add command option to split voice into multiple channels?

  • voice-tone
  • voice-tone-roughness
  • consonants voiced
  • consonants unvoiced
  • breath

(time/pitch stretching, noise removal, compression, remastering)


Updates to Github #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng:master] New Issue Created by cmrdt:
#178 upgrade internal audio quality (sample rate and bit-depth)

also in espeakedit (not ng) the resampling quality seemed to be so-so (maybe not sox's fault, maybe command-line-settings) but resampling from a low-quality source-wav file isn't the same as starting from one of higher quality (lower rates lack higher frequencies)

default currently is 22050 HZ 16-bits maybe increase to 32kHz+ and 32-bit-float as compile option; & rather do downsampling from higher quality if need be (update syllables containing high frequencies: s f z ʃ ʒ t k θ ð )

and if using sox use best-quality sinc option, max precision


[espeak-ng:master] New Issue Created by cmrdt:
#178 upgrade internal audio quality (sample rate and bit-depth)

also in espeakedit (not ng) the resampling quality seemed to be so-so (maybe not sox's fault, maybe command-line-settings) but resampling from a low-quality source-wav file isn't the same as starting from one of higher quality (lower rates lack higher frequencies)

default currently is 22050 HZ 16-bits maybe increase to 32kHz+ and 32-bit-float as compile option; & rather do downsampling from higher quality if need be (update syllables containing high frequencies: s f z ʃ ʒ t k θ ð v )

and if using sox use best-quality sinc option, max precision


[espeak-ng:master] New Issue Created by cmrdt:
#178 upgrade internal audio quality (sample rate and bit-depth)

also in espeakedit (not ng) the resampling quality seemed to be so-so (maybe not sox's fault, maybe command-line-settings) but resampling from a low-quality source-wav file isn't the same as starting from one of higher quality (lower rates lack higher frequencies)

default currently is 22050 HZ 16-bits maybe increase to 32kHz+ and 32-bit-float as compile option; & rather do downsampling from higher quality if need be (update syllables containing high frequencies: s f z ʃ ʒ t k θ ð v d )

and if using sox use best-quality sinc option, max precision


Updates to Github #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng:master] New Issue Created by cmrdt:
#178 upgrade internal audio quality (sample rate and bit-depth)

also in espeakedit (not ng) the resampling quality seemed to be so-so (maybe not sox's fault, maybe command-line-settings) but resampling from a low-quality source-wav file isn't the same as starting from one of higher quality (lower rates lack higher frequencies)

default currently is 22050 HZ 16-bits maybe increase to 32kHz+ and 32-bit-float as compile option; & rather do downsampling from higher quality if need be (update syllables containing high frequencies: s f z ʃ ʒ t )

and if using sox use best-quality sinc option, max precision


[espeak-ng:master] New Issue Created by cmrdt:
#178 upgrade internal audio quality (sample rate and bit-depth)

also in espeakedit (not ng) the resampling quality seemed to be so-so (maybe not sox's fault, maybe command-line-settings) but resampling from a low-quality source-wav file isn't the same as starting from one of higher quality (lower rates lack higher frequencies)

default currently is 22050 HZ 16-bits maybe increase to 32kHz+ and 32-bit-float as compile option; & rather do downsampling from higher quality if need be (update syllables containing high frequencies: s f z ʃ ʒ t k )

and if using sox use best-quality sinc option, max precision


[espeak-ng:master] New Issue Created by cmrdt:
#178 upgrade internal audio quality (sample rate and bit-depth)

also in espeakedit (not ng) the resampling quality seemed to be so-so (maybe not sox's fault, maybe command-line-settings) but resampling from a low-quality source-wav file isn't the same as starting from one of higher quality (lower rates lack higher frequencies)

default currently is 22050 HZ 16-bits maybe increase to 32kHz+ and 32-bit-float as compile option; & rather do downsampling from higher quality if need be (update syllables containing high frequencies: s f z ʃ ʒ t k θ )

and if using sox use best-quality sinc option, max precision


[espeak-ng:master] new issue: upgrade internal audio quality (sample rate and bit-depth) #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng:master] New Issue Created by cmrdt:
#178 upgrade internal audio quality (sample rate and bit-depth)

also in espeakedit (not ng) the resampling quality seemed to be so-so (maybe not sox's fault, maybe command-line-settings) but resampling from a low-quality source-wav file isn't the same as starting from one of higher quality (lower rates lack higher frequencies)

default currently is 22050 HZ 16-bits maybe increase to 32kHz+ and 32-bit-float as compile option; & rather do downsampling from higher quality if need be

and if using sox use best-quality sinc option, max precision


[espeak-ng:master] new issue: upgrade internal audio quality (sample rate and bit-depth) #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng:master] New Issue Created by cmrdt:
#178 upgrade internal audio quality (sample rate and bit-depth)

also in espeakedit (not ng) the resampling quality seemed to be so-so (maybe not sox's fault, maybe command-line-settings) but resampling from a low-quality source-wav file isn't the same as starting from one of higher quality

default currently is 22050 HZ 16-bits maybe increase to 32kHz+ and 32-bit-float compile option to do down-sample

and if using sox use best-quality sinc option, max precision


Updates to Github #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng:master] New Issue Created by cmrdt:
#178 upgrade internal audio quality (sample rate and bit-depth)

also in espeakedit (not ng) the resampling quality seemed to be so-so (maybe not sox's fault, maybe command-line-settings) but resampling from a low-quality source-wav file isn't the same as starting from one of higher quality

default currently is 22050 HZ 16-bits maybe increase to 32kHz+ and 32-bit-float compile option to do down-sample

and if using sox use best-quality sinc option, max precision


[espeak-ng:master] New Issue Created by cmrdt:
#179 multi-channel voice output for DSPs, mixing or debugging

add command option to split voice into multiple channels?

  • voice-tone
  • voice-tone-roughness
  • consonants voiced
  • consonants unvoiced
  • breath

(time/pitch stretching, noise removal, compression)


[espeak-ng:master] New Comment on Issue #178 upgrade internal audio quality (sample rate and bit-depth)
By cmrdt:

also if audio file same quality as device native-sampling rate, it would not have to resample again


[espeak-ng:master] new issue: upgrade internal audio quality (sample rate and bit-depth) #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng:master] New Issue Created by cmrdt:
#178 upgrade internal audio quality (sample rate and bit-depth)

also in espeakedit (not ng) the resampling quality seemed to be so-so (maybe not sox's fault, maybe command-line-settings) but resampling from a low-quality source-wav file isn't the same as starting from one of higher quality

default currently is 22050 HZ 16-bits maybe increase to 32kHz+ and 32-bit-float compile option to do down-sample

and if using sox use best-quality sinc option, max percision


Updates to Github #github

espeak-ng@groups.io Integration <espeak-ng@...>
 

[espeak-ng:master] New Issue Created by rhdunn:
#177 Merge the android branch into master.

Now that espeak-ng has diverged from espeak, it makes sense to have the android branch merged into the main development line. This will make it easier to maintain the Android support in the future and keep it up-to-date. To facilitate this, the Android targets should be merged into the makefile, supporting something like make apk.

The Android-specific change should be implemeted using the current compat structure for cross-platform compatibility. This includes compatibility support for the wide-character functions using the ucd-tools project. NOTE: The ucd-tools project should be included using a git subtree.

The README should be combined, with the relevant parts incorporated.


[espeak-ng:master] Label added to issue #177 Merge the android branch into master. by rhdunn.


[espeak-ng:master] Label added to issue #177 Merge the android branch into master. by rhdunn.


Re: Updated Wikipedia article about eSpeakNG

abdelwaheb turki <turkiabdelwaheb@...>
 

Dear Sir,

I thank you for your answer. The work is better. The history part is well written now by Mr. Reece. Sure, you can add a list of supported languages with their ISO 639-3 codes to the official page of eSpeakNG and use it as a reference for languages part in Wikipedia. This will be better than picking the languages from many references.  For the features and principles parts, you can use research papers like https://www.researchgate.net/profile/Anto_Nugroho/publication/261244975_Evaluation_of_text-to-speech_synthesizer_for_Indonesian_language_using_semantically_unpredictable_sentences_test_IndoTTS_eSpeak_and_google_translate_TTS/links/0deec5340bc0bc425c000000.pdf and http://dare.uva.nl/cgi/arno/show.cgi?fid=494975... If there is any problem, please write me and I will solve the problem. However, I think that the work will be kept in Wikipedia. 

Yours Sincerely,

Houcemeddine Turki


De : espeak-ng@groups.io <espeak-ng@groups.io> de la part de Reece H. Dunn <msclrhd@...>
Envoyé : mardi 27 décembre 2016 13:17
À : espeak-ng@groups.io
Objet : Re: [espeak-ng] Updated Wikipedia article about eSpeakNG
 
On Tue, Dec 27, 2016 at 03:34 am, Valdis Vitolins wrote:
I got list of languages, at first with:

espeak-ng --voices|grep -vE "Language"|awk '{print $2" "$4}'|sort -u

but, to filer out just dialects etc., I used following Bash script in root folder of the eSpeakNG project:

#!/bin/bash
cd dictsource
list=$(ls -1 *_rules|cut -d'_' -f1)
cd ..
for i in $list; do
  langfile=$(find espeak-ng-data -name $i)
  if [ -n "$langfile" ]; then 
    lang=$(cat $langfile|grep name|tr "\-+" "  "| cut -d' ' -f2)
    echo $i $lang
  fi
done

Can't we use reference to the source, e.g.: https://github.com/espeak-ng/espeak-ng/tree/master/dictsource?

Reece, to prove eSpeakNG's capabilities, maybe we need to add list of supported languages for eSpeakNG,

espeak.sourceforge.net
Help Needed Many of these are just experimental attempts at these languages, produced after a quick reading of the corresponding article on wikipedia.org.


There is http://reecedunn.co.uk/espeak-for-android for my Android port, but that is not an up-to-date list. I will add the list of supported languages to the README file.

reecedunn.co.uk
The eSpeak for Android program is a port of the eSpeak Text-to-Speech synthesiser to Android, supporting 79 languages and accents. This port of eSpeak is built on the ...




I don't know about any sources of eSpeak history, but I could add some more detailed explanation about principles later.

I am currently writing the history section of the Wikipedia article :).


Kind redards,

Reece


Re: Updated Wikipedia article about eSpeakNG

Reece H. Dunn
 

On Tue, Dec 27, 2016 at 03:34 am, Valdis Vitolins wrote:
I got list of languages, at first with:

espeak-ng --voices|grep -vE "Language"|awk '{print $2" "$4}'|sort -u

but, to filer out just dialects etc., I used following Bash script in root folder of the eSpeakNG project:

#!/bin/bash
cd dictsource
list=$(ls -1 *_rules|cut -d'_' -f1)
cd ..
for i in $list; do
  langfile=$(find espeak-ng-data -name $i)
  if [ -n "$langfile" ]; then 
    lang=$(cat $langfile|grep name|tr "\-+" "  "| cut -d' ' -f2)
    echo $i $lang
  fi
done

Can't we use reference to the source, e.g.: https://github.com/espeak-ng/espeak-ng/tree/master/dictsource?

Reece, to prove eSpeakNG's capabilities, maybe we need to add list of supported languages for eSpeakNG,

There is http://reecedunn.co.uk/espeak-for-android for my Android port, but that is not an up-to-date list. I will add the list of supported languages to the README file.

I don't know about any sources of eSpeak history, but I could add some more detailed explanation about principles later.

I am currently writing the history section of the Wikipedia article :).


Kind redards,

Reece


Re: Updated Wikipedia article about eSpeakNG

Valdis Vitolins
 

I got list of languages, at first with:

espeak-ng --voices|grep -vE "Language"|awk '{print $2" "$4}'|sort -u

but, to filer out just dialects etc., I used following Bash script in root folder of the eSpeakNG project:

#!/bin/bash
cd dictsource
list=$(ls -1 *_rules|cut -d'_' -f1)
cd ..
for i in $list; do
  langfile=$(find espeak-ng-data -name $i)
  if [ -n "$langfile" ]; then 
    lang=$(cat $langfile|grep name|tr "\-+" "  "| cut -d' ' -f2)
    echo $i $lang
  fi
done

Can't we use reference to the source, e.g.: https://github.com/espeak-ng/espeak-ng/tree/master/dictsource?

Reece, to prove eSpeakNG's capabilities, maybe we need to add list of supported languages for eSpeakNG,

I don't know about any sources of eSpeak history, but I could add some more detailed explanation about principles later.

Valdis

Dear Sir,

I adjusted the list of languages. However, I did not found a reference from several languages. As well, I found references for supported languages that you did not include in the list. So, please verify this part before I can add references to the other parts of the work. As well, please expand the other parts by including further details about the history and the principles of eSpeakNG.

Yours Sincerely,

Houcemeddine Turki


De : espeak-ng@groups.io <espeak-ng@groups.io> de la part de abdelwaheb turki <turkiabdelwaheb@...>
Envoyé : mardi 27 décembre 2016 10:01
À : espeak-ng@groups.io
Objet : Re: [espeak-ng] Updated Wikipedia article about eSpeakNG
 

Dear Sir,

I thank you for your answer. It is better now. It can pass. I will add references to it.

Yours Sincerely,

Houcemeddine Turki


De : espeak-ng@groups.io <espeak-ng@groups.io> de la part de Valdis Vitolins <valdis.vitolins@...>
Envoyé : mardi 27 décembre 2016 09:09
À : espeak-ng@groups.io
Objet : Re: [espeak-ng] Updated Wikipedia article about eSpeakNG
 
Hello, Houcemeddine!

I've corrected list of supported languages by eSpeakNG, because previously I used list of supported voices, which are related to languages, but
sometimes are not the same.

I have no idea where we can get more prominent sources for references.
All I can say is that current page has some more references than before, so, if quality criteria are not risen much higher, it should pass.

All the best from me too.

Valdis

Dear Sir,

I did not tell you that I am one of the 2000 most contributing users in the English Wikipedia. I am a founding member of Wikimedia TN, Wikipedia Library, WikiProject Maghrebi Arabic, and Wikimedia Maghreb. My entries was featured four times in the main page of English Wikipedia. I must tell you that the work lacks of references. It includes some misleading information. For example, I ask if Mohamed and Brazil are languages. As well, blog posts are unfortunately not considered as a trustworthy reference by the editors of Wikipédia. It can be removed by one week if someone does not add references to it. Please send me references (journal items, researches...) about eSpeak. I will edit the work. I congratulate you for the work and I invite you to include Arabic dialects to it in the next years. Merry Christmas and Happy New Year 2017.

Yours Sincerely,

Houcemeddine Turki


De : espeak-ng@groups.io <espeak-ng@groups.io> de la part de Valdis Vitolins <valdis.vitolins@...>
Envoyé : dimanche 25 décembre 2016 17:26
À : espeak-ng@groups.io
Objet : [espeak-ng] Updated Wikipedia article about eSpeakNG
 
After loosing battle with bureaucratic bots and moderators with two
Wikipedia articles: eSpeak and eSpeakNG,

I renamed original eSpeak article to eSpeakNG and updated it, to
reflect relevant information about eSpeakNG project.

https://en.wikipedia.org/wiki/ESpeakNG

Reviews and updates are welcome.

Valdis