[Freeswitch-users] Problem with personalized locutions and use of say with type shor_date_time.

Miguel Jesús López Valverde mjlopez at smartic.es
Thu Dec 14 11:53:41 UTC 2017


Hello Anatoli, thank you very much for your answer.

 

                At this time, the sets of locutions in Neutral-Spanish that I have available have been obtained with a TTS only with laboratory interest. I do not consider them suitable for a more professional or commercial use.

 

On the other hand, we are evaluating making recordings through some Spanish studies that have made us offers, based on the list available by default of locutions in English. For your information, the prices are approx. about 1800 Euros per group of locutions, (initially we would only make a group in female voice) considering that the studies that have offered us maintain a high level of quality in our point of view, since they have already done other work for us with optimal results .

 

Please, indicate if you have a more complete list than the one used by us to facilitate these studies, (the list that we have handled has been the one of locutions in English given in the version 1.6.18), so that we can pass it to assess.

 

In case of getting to make the recordings of the locutions, we could assess some form of cooperation to facilitate them to the community, (shared payment, training agreements or certification courses, etc ..).

 

Please, indicate if you are interested because, in this case, it may also be of interest to us to contract the completion of the locutions.

 

Again, Thank you very much and receive greeting.

 

 

 

De: FreeSWITCH-users [mailto:freeswitch-users-bounces at lists.freeswitch.org] En nombre de Anatoli
Enviado el: jueves, 14 de diciembre de 2017 3:02
Para: freeswitch-users at lists.freeswitch.org; consulting at freeswitch.org
Asunto: Re: [Freeswitch-users] Problem with personalized locutions and use of say with type shor_date_time.

 

Hi Miguel Jesús,

That's because mod_say_es doesn't implement all say types, but most of the TTS configs use them (e.g. lang/en/vm/sounds.xml:396 <action function="say" data="$1" method="pronounced" type="short_date_time"/>). So, if you use as the base the English configs (because the Spanish base lacks voicemail or conference configs), mod_say_es won't work.

The Spanish sounds/prompts are broken. You may try to use mod_say_es_ar which is based on the latest mod_say_en, but it has some issues with some prompts anyway (though easier to solve). Basically, the problem is that the mod_say_es is outdated, the TTS configs are not in sync with it and the available Spanish prompts (sounds_es_mx_maria and sounds_es_ar_mario, I couldn't find others) are also not in sync with the configs and the say modules, and are incomplete (e.g. no "un" sound for "usted tiene un mensaje nuevo" ("you have one new message"), some words lack singular or plural forms, some prompts are completely absent (e.g. "entered the conference"), etc.).

I had to mix sounds from the 2 sets (MX and AR) and even borrow some sounds from the asterisk sounds. Even this way the result is not great (masculine and feminine voices mixed in the same sentence, lack of singular/plural forms for some words, etc.), but at least it works and the users can interact with the system.

The Spanish support in FS is not production ready. What should be done?

1.	mod_say_es should be updated to be in sync and as complete as the rest of the say modules (should be based on mod_say_en as this is the reference). And there is NO need for country-specific Spanish *say modules*, i.e. the grammar (the sentence composition) is the same in all regions. Only the prompts should be recorded by native speakers of each country.
2.	The Spanish TTS configs should be updated accordingly (again, NO country-specific configs).
3.	The list of necessary sounds (i.e. the prompts (texts) to country-adapt and record) should be defined (either in English or in neutral Spanish).
4.	A country-neutral set of Spanish sounds should be created.

Once this is ready, the community could provide the prompts for country-specific dialects. And I'm ready to co-sponsor the Spanish base (the 4 items list above) and to provide the correct es_AR prompts. If someone can implement the Spanish base (at least the first 3 items above), please let us know your price, I guess we could crowd-fund it.

Regards,
Anatoli

From: Miguel Jesús López Valverde
Sent: Wednesday, December 13, 2017 14:51
To: Freeswitch-users
Subject: [Freeswitch-users] Problem with personalized locutions and use of say with type shor_date_time.

 

Hello Freeswitch users:

 

I send this email to you with a new problem found.

 

I have a set of complete locutions created with a TTS for Spanish locutions installed in Freeswitch, along with those brought by default in English.

 

When using the phrases in Spanish, I am only having problems when using the directive:

<action function = "say" data = "$ 1" method = "pronounced" type = "short_date_time" />

but with other directives "say" with different methods, I'm not having any problems. For example, when accessing voicemail using these locutions set in Spanish, I get the following error that prevents access to recorded messages:

 

2017-12-13 16:58:52.252852 [DEBUG] switch_ivr_play_say.c:250 Handle play-file:[voicemail/vm-new.wav] (es_CB:es)

2017-12-13 16:58:52.252852 [DEBUG] switch_ivr_play_say.c:1498 Codec Activated L16 at 8000hz 1 channels 20ms

2017-12-13 16:58:53.172924 [DEBUG] switch_ivr_play_say.c:1942 done playing file /usr/local/freeswitch/sounds/es/colabora/sonia/voicemail/vm-new.wav

2017-12-13 16:58:53.292911 [DEBUG] switch_ivr_play_say.c:250 Handle play-file:[voicemail/vm-message_number.wav] (es_CB:es)

2017-12-13 16:58:53.292911 [DEBUG] switch_ivr_play_say.c:1498 Codec Activated L16 at 8000hz 1 channels 20ms

2017-12-13 16:58:54.832899 [DEBUG] switch_ivr_play_say.c:1942 done playing file /usr/local/freeswitch/sounds/es/colabora/sonia/voicemail/vm-message_number.wav

2017-12-13 16:58:54.932903 [DEBUG] switch_ivr_play_say.c:250 Handle say:[1] (es_CB:es)

2017-12-13 16:58:54.932903 [DEBUG] switch_ivr_play_say.c:1498 Codec Activated L16 at 8000hz 1 channels 20ms

2017-12-13 16:58:55.872903 [DEBUG] switch_ivr_play_say.c:1942 done playing file /usr/local/freeswitch/sounds/es/colabora/sonia/digits/1.wav

2017-12-13 16:58:55.972959 [DEBUG] switch_ivr_play_say.c:70 No language specified - Using [es_CB]

2017-12-13 16:58:55.972959 [DEBUG] switch_ivr_play_say.c:250 Handle say:[1513180033] (es_CB:es)

2017-12-13 16:58:55.972959 [ERR] mod_say_es.c:471 Unknown Say type=[18]

 

By changing this language by default English, it works correctly ...

 

2017-12-13 17:18:05.327117 [DEBUG] switch_ivr_play_say.c:70 No language specified - Using [en]

2017-12-13 17:18:05.367151 [DEBUG] switch_ivr_play_say.c:250 Handle play-file:[voicemail/vm-new.wav] (en:en)

2017-12-13 17:18:05.367151 [DEBUG] switch_ivr_play_say.c:1498 Codec Activated L16 at 8000hz 1 channels 20ms

2017-12-13 17:18:05.667117 [DEBUG] switch_ivr_play_say.c:1942 done playing file /usr/local/freeswitch/sounds/en/us/callie/voicemail/vm-new.wav

2017-12-13 17:18:05.767136 [DEBUG] switch_ivr_play_say.c:250 Handle play-file:[voicemail/vm-message_number.wav] (en:en)

2017-12-13 17:18:05.767136 [DEBUG] switch_ivr_play_say.c:1498 Codec Activated L16 at 8000hz 1 channels 20ms

2017-12-13 17:18:06.587140 [DEBUG] switch_ivr_play_say.c:1942 done playing file /usr/local/freeswitch/sounds/en/us/callie/voicemail/vm-message_number.wav

2017-12-13 17:18:06.707138 [DEBUG] switch_ivr_play_say.c:250 Handle say:[1] (en:en)

2017-12-13 17:18:06.707138 [DEBUG] switch_ivr_play_say.c:1498 Codec Activated L16 at 8000hz 1 channels 20ms

2017-12-13 17:18:07.407128 [DEBUG] switch_ivr_play_say.c:1942 done playing file file_string://digits/1.wav

2017-12-13 17:18:07.507133 [DEBUG] switch_ivr_play_say.c:70 No language specified - Using [en]

2017-12-13 17:18:07.527108 [DEBUG] switch_ivr_play_say.c:250 Handle say:[1513180033] (en:en)

2017-12-13 17:18:07.527108 [DEBUG] switch_ivr_play_say.c:1498 Codec Activated L16 at 8000hz 1 channels 20ms

2017-12-13 17:18:10.427138 [DEBUG] switch_ivr_play_say.c:1942 done playing file file_string://time/today.wav!time/at.wav!digits/3.wav!digits/40.wav!digits/7.wav!time/p-m.wav

2017-12-13 17:18:10.547147 [DEBUG] switch_ivr_play_say.c:1498 Codec Activated L16 at 8000hz 1 channels 20ms

 

It is proven that I have all the locutions in Spanish, including those in the "time" and "digits" folders.

 

Does anyone know how I can get a proper voicemail operation with personalized locutions when using the function "say" with "type =" short_date_time "" ?.

 

Thank you very much and best regards.

 

 


 <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient> 

Libre de virus.  <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient> www.avast.com 






_________________________________________________________________________
Professional FreeSWITCH Consulting Services:
consulting at freeswitch.org <mailto:consulting at freeswitch.org> 
http://www.freeswitchsolutions.com
 
Official FreeSWITCH Sites
http://www.freeswitch.org
http://confluence.freeswitch.org
http://www.cluecon.com
 
FreeSWITCH-users mailing list
FreeSWITCH-users at lists.freeswitch.org <mailto:FreeSWITCH-users at lists.freeswitch.org> 
http://lists.freeswitch.org/mailman/listinfo/freeswitch-users
UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users
http://www.freeswitch.org

 



---
El software de antivirus Avast ha analizado este correo electrónico en busca de virus.
https://www.avast.com/antivirus
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freeswitch.org/pipermail/freeswitch-users/attachments/20171214/6c779c76/attachment-0001.html>


More information about the FreeSWITCH-users mailing list