Date   

ESPEAK IN MICROSOFT?

Ricky Lomey
 

Hi

Even though Narrator and Espeak are of course totally different products and
comparisons are odious, is there a way to add Espeak to Narrator or parts of
Espeak to Narrator as a voice in some way, say to add Afrikaans as a
language in Narrator or any other it doesn't use and then to add Espeak or
use Espeak somehow else in Microsoft or can it only be used with Window-Eyes
or Jaws or perhaps NVDA because I just find it so much easier to switch
languages using Espeak.

Thanks.

Ricky Lomey


Pull Request Opened #github

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

[espeak-ng/espeak-ng] Pull request opened by guest271314:

#743 Fix link to tts:style


[espeak-ng:master] reported: Cracks when voicing multi-line text #github

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

[espeak-ng:master] New Comment on Issue #742 Cracks when voicing multi-line text
By SeanTolstoyevski:

No no, this is not what I want to do.

I can delete the lines myself.

I need a parameter for Espeak to ignore the lines.

But still thank you.


[espeak-ng:master] reported: Cracks when voicing multi-line text #github

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

[espeak-ng:master] New Comment on Issue #742 Cracks when voicing multi-line text
By guest271314:

I don't know what is SSML.

See https://www.w3.org/TR/2010/REC-speech-synthesis11-20100907/. For example essentially an opposite requirement to what is described at this issue https://github.com/espeak-ng/espeak-ng/issues/739.

With which parameter can I remove the pause between lines?

To be sure, I need to remove the pause between lines, because NVDA does not pause between lines.

Have not thus far located any espeak-ng option to substitute multiple space characters and/or newline characters for single space character.

--

Potential workarounds:

Am not able to test at a *indows OS.

At *nix using gedit https://wiki.gnome.org/Apps/Gedit Find and Replace with a Regular expression can be used to replace \s+ with .

Screenshot_2020-05-03_20-43-39

A local HTML file can also be used to replace \s+ with a single space character.

<!doctype html>

<html>
  <head>
  </head>

  <body>
    <input type="file">
    <script>
      const input = document.querySelector('input[type=file]');
      input.onchange = e => {
        const {files:[file]} = e.target;
        const reader = new FileReader;
        reader.onload = _ => {
          const text = reader.result.replace(/\s+/g, ' ');
          window.open(`data:application/octet-stream,${encodeURIComponent(text)}`);
        }
        reader.readAsText(file);
      }
    </script>
  </body>
</html>

results in a file downloaded to local filesystem with \s+ replaced with single space character .

"test 123 Test one test two Table of content Part 1 : a night part 2 : african part 3 : the end world 


[espeak-ng:master] reported: Cracks when voicing multi-line text #github

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

[espeak-ng:master] New Comment on Issue #742 Cracks when voicing multi-line text
By SeanTolstoyevski:

I did not add the wav file because I could not create the problem in NVDA in Espeak. I don't know what is SSML. With which parameter can I remove the pause between lines?

Also, when you test it with this file, you will see that it is okay. To be sure, I need to remove the pause between lines, because NVDA does not pause between lines.


[espeak-ng:master] reported: Cracks when voicing multi-line text #github

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

[espeak-ng:master] New Comment on Issue #742 Cracks when voicing multi-line text
By guest271314:

The .zip file only contains plain text not audio output as a wav file.

Does NVDA support SSML input?


[espeak-ng:master] reported: Cracks when voicing multi-line text #github

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

[espeak-ng:master] New Comment on Issue #742 Cracks when voicing multi-line text
By SeanTolstoyevski:

espeak-test.zip this is txt file. Test file

Hi,

There is a difference. Espeak pauses between lines.

NVDA does not pause in this type of text. it reads it as a plain sentence.

I did not see any problem when I tested Espeak.

By the way, this happens when the rate of sound is too high. So this problem does not occur at normal rate.

If there is a parameter I can remove the pauses between lines, I would like to try again. I read --help but couldn't find it. The command line parameters I use are:

espeak-ng.exe -f espeak-test.txt -w output.wav -s 450 -g 0

If the problem is not in Espeak, we will continue to topic this issue in NVDA. As a note: I can say that there are more sound cracks in Turkish. I can not decide because I am unable to create the same environment at the moment.


[espeak-ng:master] reported: Cracks when voicing multi-line text #github

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

[espeak-ng:master] New Comment on Issue #742 Cracks when voicing multi-line text
By guest271314:

You can save the file as WAV

$ espeak-ng -f input.txt -w output.wav

where input.txt is the text printed in linked issue saved in a file then save the file in a .zip archive then upload here within a comment.


Updates to Github #github

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

[espeak-ng:master] New Comment on Issue #742 Cracks when voicing multi-line text
By guest271314:

You can either save the file as WAV

$ espeak-ng -w output.wav 'test'

or at Firefox, record the output using MediaRecorder, see https://github.com/web-platform-tests/wpt/issues/23084#issuecomment-616151476, then save the file in a .zip archive then upload here within a comment.


[espeak-ng:master] New Comment on Issue #742 Cracks when voicing multi-line text
By guest271314:

You can either save the file as WAV

$ espeak-ng -w output.wav 'test'

then save the file in a .zip archive then upload here within a comment.


[espeak-ng:master] New Comment on Issue #742 Cracks when voicing multi-line text
By guest271314:

You can either save the file as WAV

$ espeak-ng -f input.txt -w output.wav 'test'

where input.txt is the text printed in linked issue saved in a file then save the file in a .zip archive then upload here within a comment.


[espeak-ng:master] New Comment on Issue #742 Cracks when voicing multi-line text
By guest271314:

You can either save the file as WAV

$ espeak-ng -f input.txt -w output.wav

where input.txt is the text printed in linked issue saved in a file then save the file in a .zip archive then upload here within a comment.


[espeak-ng:master] reported: Cracks when voicing multi-line text #github

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

[espeak-ng:master] New Comment on Issue #742 Cracks when voicing multi-line text
By SeanTolstoyevski:

I don't know how I can do it.

But I will record sound.


[espeak-ng:master] reported: Does this codecs compatible with espeakNG ? #github

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

[espeak-ng:master] New Comment on Issue #724 Does this codecs compatible with espeakNG ?
By guest271314:

It is not clear what the plugin does.

espeak-ng outputs audio encoded as audio/wav.

Does the plugin pipe default audio output from espeak and/or espeak-ng through a native application to output a different, user-selected audio codec?

If that is the case an intermediary application is not necessary, for example, to convert default audio/wav output stream to audio/ogg;codecs=opus (audio/x-opus+ogg)

$ espeak-ng --stdout 'test' | opusenc - -


[espeak-ng:master] reported: Cracks when voicing multi-line text #github

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

[espeak-ng:master] New Comment on Issue #742 Cracks when voicing multi-line text
By guest271314:

Can you record the output and post here?


[espeak-ng:master] new issue: Cracks when voicing multi-line text #github

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

[espeak-ng:master] New Issue Created by SeanTolstoyevski:
#742 Cracks when voicing multi-line text

Hi espeak's developers,

This problem was first created in NVDA.

The problem is thought to be in Espeak.

Please check.

Here is the issue: https://github.com/nvaccess/nvda/issues/11061


[espeak-ng:master] new issue: Spurious punctuation pronunciation #github

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

[espeak-ng:master] New Issue Created by sthibaul:
#741 Spurious punctuation pronunciation

Hello,

Running espeak-ng 1.50:

espeak-ng -v en-us -m '<speak><tts:style field="punctuation" mode="none">autogen.sh.</tts:style></speak>'

pronounces “autogen dot sh dot”. mode="some" and mode="most" get the same result.

This is a regression from 1.49, with which we get “autogen dot sh”.

This happens also with "-v es" for instance. This doesn't happen with "-v fr", "-v it", or "-v de".

Samuel


Pull Request Updated #github

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

[espeak-ng/espeak-ng] Pull request updated by BenTalagan:

#740 Added pronunciation of the name 'Tolkien' and derivatives

Hi, this is a mini-PR to handle the name of the author of the LotR (and derivatives : tolkienian, tolkienic, tolkiendil, etc) in english.


Updates to Github #github

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

[espeak-ng/espeak-ng] Pull request updated by BenTalagan:

#740 Added pronounciation of the name Tolkien and derivatives

Hi, this is a mini-PR to handle the name of the LotR's author (and derivatives : tolkienian, tolkienic, tolkiendil, etc) in english.


[espeak-ng/espeak-ng] Pull request updated by BenTalagan:

#740 Added pronounciation of the name Tolkien and derivatives

Hi, this is a mini-PR to handle the name of the author of the LotR (and derivatives : tolkienian, tolkienic, tolkiendil, etc) in english.


[espeak-ng/espeak-ng] Pull request updated by BenTalagan:

#740 Added pronunciation for the name 'Tolkien' and derivatives

Hi, this is a mini-PR to handle the name of the author of the LotR (and derivatives : tolkienian, tolkienic, tolkiendil, etc) in english.


Pull Request Opened #github

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

[espeak-ng/espeak-ng] Pull request opened by BenTalagan:

#740 Added pronounciation of the name Tolkien and derivatives

Hi, this is a mini-PR to handle the name of the LotR author (and derivatives : tolkienian, tolkienic, tolkiendil, etc) in english.


[espeak-ng:master] closed: Update for Turkish#github

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

[espeak-ng:master] Issue #734 Update for Turkish closed by boracasli98.


Updates to Github #github

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


[espeak-ng:master] reported: How to change default voice #github

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

[espeak-ng:master] New Comment on Issue #719 How to change default voice
By guest271314:

Note, installing a modified version of espeak-ng, e.g., following the instructions for building espeak-ng does not result in those changes being reflected at /usr/lib/speech-dispatcher-modules/sd_espeak-ng, in part

libespeak-ng.so.1_ITM_deregisterTMCloneTable__gmon_start___ITM_registerTMCloneTableespeak_SetVoiceByNameespeak_Initializeespeak_SetSynthCallbackespeak_Infoespeak_SetPunctuationListespeak_SetUriCallbackespeak_ListVoicesespeak_SetParameterespeak_Synthespeak_Cancelespeak_Terminatelibsndfile.so.1sf_commandsf_opensf_strerrorsf_read_shortsf_closelibdotconf.so.0dotconf_createdotconf_cleanupdotconf_command_looplibglib-2.0.so.0g_strndupg_string_appendg_malloc0_ng_string_newg_utf8_strchrg_strdelimitg_str_has_prefixg_strdupg_strfreevg_strstr_leng_queue_peek_head_linkg_queue_freeg_malloc0g_string_append_printfg_reallocg_strsplitg_utf8_skipg_memdupg_queue_push_tailg_convertg_queue_free_fullg_convert_with_fallbackg_strcmp0g_string_freeg_utf8_validateg_mallocg_slist_appendg_file_testg_queue_get_lengthg_queue_is_emptyg_strdup_printfg_queue_newg_slist_removeg_usleeplibltdl.so.7lt_dlexitlt_dlsymlt_dlcloselt_dladvise_initlt_dlsetsearchpathlt_dlpreload_defaultlt_dladvise_destroylt_dladvise_extlt_dladvise_globallt_dlopenadviselt_dlerrorlt_dlinitlibpthread.so.0pthread_cond_destroypthread_createpthread_cancelpthread_joinpthread_setcanceltypepthread_mutex_initpthread_setcancelstatepthread_cond_broadcastsem_initpthread_cond_waitpthread_mutex_unlockpthread_cond_signalsem_postpthread_sigmasksem_trywaitpthread_cond_init__errno_locationsem_destroypthread_mutex_lockpthread_mutex_trylockpthread_mutex_destroypthread_exitsem_waitlibc.so.6_IO_stdin_usedfflush__printf_chkfopenstrncmpsigprocmasksigfillsetstdin__assert_failstrtokstrtolstrlenungetc__fprintf_chksigaddsetctimestdoutfputcfputsfclosestderrsigdelsetfwritegettimeofday__cxa_finalize_IO_getcstrcmp__libc_start_main__stack_chk_fail_edata__bss_startmodule_stopmodule_send_asynchronousmodule_recode_to_isomodule_parent_dp_readmodule_index_markcurrent_index_markmodule_strip_ssmlstr2ESpellModemodule_initmodule_sigunblockusrspd_audio_opendo_charmodule_parent_wforkmodule_list_voicesEspeakMaxRate_cb_fp_hwmodule_report_event_endstr2intpriorityEspeakPunctuationList_cbmodule_strip_punctuation_someEPunctMode2strmodule_report_event_stopdo_speakdo_stopmodule_play_filedo_debugmodule_report_event_begindo_messageDebugspd_getlinespd_audio_set_logleveldo_pausedo_audiomodule_parent_dp_initlt__PROGRAM__LTX_preloaded_symbolsEspeakMinRate_cbmodule_child_dp_readmodule_report_index_markmodule_sigblockallconfigfiledo_setspd_audio_set_volumemodule_audio_idmodule_speakmodule_tts_outputstr2EPunctModemodule_terminate_threadmodule_pausemodule_parent_dp_closemodule_child_dp_writeESpellMode2strstr2ECapLetRecognmsg_settingsdo_list_voicesspd_audio_stopspd_audio_closemodule_child_dp_initdo_sound_icon__data_startdo_loglevelmodule_strip_punctuation_defaultmodule_parent_dp_writeplayback_queue_conditionmodule_child_dp_closeEVoice2strECapLetRecogn2strCustomDebugFilelog_level__libc_csu_initEspeakNormalRate_cbEspeakSoundIconVolume_cbEspeakCapitalPitchRise_cbdo_quitstr2EVoicemodule_audio_initmodule_parent_wait_continueEspeakAudioQueueMaxSize_cbmsg_settings_oldmodule_num_dc_optionsspd_audio_get_playcmdset_speaking_thread_parametersmodule_stdout_mutex__libc_csu_finimodule_add_config_optionmodule_sigblockusrspd_audio_playmodule_loaddo_keyEspeakSoundIconFolder_cbEspeakAudioChunkSize_cbmodule_dc_optionsmodule_report_event_pauseEspeakListVoiceVariants_cbmodule_get_message_partmodule_closeDebug_cblibsndfile.so.1.0GLIBC_2.1GLIBC_2.0GLIBC_2.3.2GLIBC_2.4GLIBC_2.1.3GLIBC_2.3.4

For example, in an attempt to set SSML parsing on by default for Web Speech API so that --enable-speech-dispatcher flag at Chromium will use the modified version of espeak-ng, removed all espeak and espeak-ng packages, added | espeakSSML to L344 at espeak-ng.c, installed using make, tested espeak-ng <speak>test</speak> at terminal which output audio "test" without passing -m flag.

$ spd-say <speak>test</test> does not output the same audio, SSML is not enabled without x option, sd_espeak-ng appears to be an entirely different version of espeak-ng unrelated to installed espeak-ng by packages.

Installing python3-speechd then running spd-conf creates ~/.config/speech-dispatcher/speechd.conf where there are fields available for setting # ----- VOICE PARAMETERS -----, DefaultVoiceType, DefaultLanguage. ~/.config/speech-dispatcher/modules/espeak-ng.conf contains fields for setting EspeakListVoiceVariants.

1201 - 1220 of 4289