[Freeswitch-users] Fwd: FreeSWITCH TTS Voice Prompt Generator

Wed Mar 27 21:06:24 MSK 2013

Here are the quote's I have so far;

French (Male, Native, French accent) - $125
German (Male, Fluent, German accent) - $500
Dutch (Male, Native, Dutch accent) - $190
English (Female, Native, American accent) - $85
English (Female, Native, British accent) - $120

Each one includes;

* 3500 words
* Full copyright/ownership/rights over the audio files
* High quality, raw recordings with a professional mic
* Translation from English script (where applicable)
* Small sample provided before hand

Therefore, I think the next steps would be to determine;

* Which languages should be included, so we can determine what the cost
would be
* List of phrases/sentences that we want recorded (are we going to use
phrase fragments to construct a sentence, or full paragraphs?)

I've looked around for a list of typical phrases, but didn't have much luck.

Any thoughts?

Cal

On Tue, Mar 26, 2013 at 7:36 PM, Michael Collins <msc at freeswitch.org> wrote:

> Awesome!
>
>
> On Tue, Mar 26, 2013 at 11:29 AM, Cal Leeming [Simplicity Media Ltd] <
> cal.leeming at simplicitymedialtd.co.uk> wrote:
>
>> It looks like there was some previous discussion/bounty on this as well;
>>
>> http://wiki.freeswitch.org/wiki/Bounty#Record_Sound_Prompts_In_Other_Languages
>>
>> The original estimate was $1200 for a single language, but I think we can
>> get MUCH lower than that, maybe 100$ per language for the voice talent.
>>
>> I'll generate a list of the prompts, get some quotes from fiverr, and see
>> what the cost would be.
>>
>> If there is enough interest in this, and we can get some donations
>> towards the cost of the voice talent, then I would be happy to donate some
>> of my own time to assist with editing, slicing, organizing etc.
>>
>> Cal
>>
>> On Tue, Mar 26, 2013 at 5:42 PM, Michael Collins <msc at freeswitch.org>wrote:
>>
>>> 3000EUR? Meh.
>>>
>>> I think you're right that something like fiverr is good for getting a
>>> basic set of sound prompts recorded or for custom work. That would be
>>> especially useful if you find a talent who can continually do updated
>>> prompts.
>>>
>>> -Michael
>>>
>>>
>>> On Tue, Mar 26, 2013 at 7:12 AM, Cal Leeming [Simplicity Media Ltd] <
>>> cal.leeming at simplicitymedialtd.co.uk> wrote:
>>>
>>>> Hi guys,
>>>>
>>>> I've just spoken with Ivona who have confirmed their pricing and
>>>> licensing conditions.
>>>>
>>>> A single developer would have to purchase a single license at 3000EUR
>>>> (valid for 1 year) to generate the voice prompt files. This license gives
>>>> us all the rights to any voice prompts that the Desktop TTS software
>>>> generates, so once an audio file is created, we then hold all the rights on
>>>> that file.
>>>>
>>>> Now, 3000EUR is a lot of money, and it would be cheaper to get a bunch
>>>> of people on fiverr.com to do real high quality voice prompts.
>>>>
>>>> Any thoughts?
>>>>
>>>> Cal
>>>>
>>>> On Thu, Mar 14, 2013 at 4:17 PM, Cal Leeming [Simplicity Media Ltd] <
>>>> cal.leeming at simplicitymedialtd.co.uk> wrote:
>>>>
>>>>> Yeah I asked them in the email to clarify what the license cost would
>>>>> be for unlimited re-distribution of TTS output.
>>>>>
>>>>> Here's an idea, slightly off-topic from TTS, but well worth
>>>>> considering, assuming these files will be static usage only, i.e. you
>>>>> generate them once and leave it.
>>>>>
>>>>> You could probably hire two German voices from Fiverr.com to do every
>>>>> one of those sentences/words for like 50 bucks... we get the majority of
>>>>> our voice talent from that site, providing they have a decent quality
>>>>> microphone, and you have some simple editing tools, then you could easily
>>>>> have a complete set within a day.
>>>>>
>>>>> You'd basically provide the voice talent with a sheet to read from,
>>>>> and specify what tone, inflection and speed you want them to use..
>>>>> repeating every word twice with a 1 second gap in between. Before you
>>>>> split, you'd throw the big file into an editing package, such as Ableton,
>>>>> and tinker around with normalization, dehiss, declick, mono, voice
>>>>> enhancement etc, until you hit the sweet spot. You can then automate the
>>>>> slicing using a simple Python script that splits the file on every 500ms of
>>>>> silence. Assuming the voice talent didn't skip a word, you can then take
>>>>> your word sheet, map this to your split files, and automatically rename
>>>>> them accordingly.
>>>>>
>>>>> Using this approach saves you a lot of time/money avoiding unnecessary
>>>>> studio work.. using a static sheet allows you to not only have automation
>>>>> of the workflow, but also means the voice talent can give an accurate cost
>>>>> (because they usually base their costs on a per word basis).. i.e. 5 bucks
>>>>> for 200 words.
>>>>>
>>>>> You could probably have an entire voice set of words/sentences of that
>>>>> size completed within a day, if you use this automation approach.
>>>>>
>>>>> Cal
>>>>>
>>>>> On Thu, Mar 14, 2013 at 3:43 PM, Julian Pawlowski <julian at pawlowski.me
>>>>> > wrote:
>>>>>
>>>>>>  On Thu, Mar 14, 2013 at 2:56 PM, Cal Leeming [Simplicity Media Ltd]
>>>>>> <cal.leeming at simplicitymedialtd.co.uk> wrote:
>>>>>>
>>>>>>> I agree their pricing is confusing for non studio related usage,
>>>>>>> I've just sent them an email asking to clarify.
>>>>>>>
>>>>>>
>>>>>> Getting back to this, after registering for the Ivona development
>>>>>> program I got access to their SaaS terms of use (
>>>>>> https://secure.ivona.com/static/pdf/saas_en.pdf).
>>>>>> It says:
>>>>>>
>>>>>> "III. SCOPE AND TYPES OF “IVONA Speech Cloud” SERVICES
>>>>>> ...
>>>>>> 3. “IVONA Speech Cloud” Service is dedicated solely to entrepreneurs.
>>>>>> Subject to the
>>>>>> provisions of these Regulations, the Service Receiver is entitled to
>>>>>> use of “IVONA
>>>>>> Speech Cloud” Service for the purposes of the business activity run
>>>>>> by Service
>>>>>> Receiver, except for business activity in the areas of Telephony
>>>>>> System, in particular
>>>>>> interactive voice response (IVR) systems, Private Automatic Branch
>>>>>> Exchange (PBX, IP
>>>>>> PABX or other) or any other telecommunication solution.
>>>>>> ...
>>>>>> IV. FREE “IVONA Speech Cloud” SERVICE
>>>>>> ...
>>>>>> 3. The Text converted into the Speech generated under Free “IVONA
>>>>>> Speech Cloud”
>>>>>> Service, will be preceded by advertising material of Ivona and/or
>>>>>> other advertising
>>>>>> material in the form of sound, to what the Service Receiver agrees
>>>>>> ordering Unpaid
>>>>>> “IVONA Speech Cloud” Service.
>>>>>> 4. The Service Receiver shall not modify, in any way, Speech
>>>>>> generated as part of Free
>>>>>> “IVONA Speech Cloud” Service.
>>>>>> 5. The Service Receiver acknowledges that the objective of provision
>>>>>> of Free “IVONA
>>>>>> Speech Cloud” Service by Ivona is primarily to enable the Service
>>>>>> Receiver to
>>>>>> familiarize with the functionality, characteristics, uses and
>>>>>> suitability of “IVONA Speech
>>>>>> Cloud” Service for the Service Receiver. Therefore, the Service
>>>>>> Receiver agrees to use
>>>>>> Speech made available to it under Free “IVONA Speech Cloud” Service
>>>>>> for the above
>>>>>> purposes only. It is prohibited to use Speech generated as part of
>>>>>> Free “IVONA Speech
>>>>>> Cloud” Service for commercial purposes, i.e. to achieve profits or
>>>>>> other material benefit
>>>>>> by the Service Receiver and/ or a third party. In particular, it is
>>>>>> prohibited to make
>>>>>> Speech available to any third parties against payment, in any manner,
>>>>>> as well as
>>>>>> reproduce, distribute, broadcast, publish Speech and on the Internet,
>>>>>> radio, television or
>>>>>> through any other media.
>>>>>> "
>>>>>>
>>>>>> That makes it impossible to use Ivona for our purposes as every
>>>>>> user/company would needs to have a valid subscription and to generate it's
>>>>>> own voice prompt files. Redistribution of pre-compiled voice files is not
>>>>>> possible.
>>>>>>
>>>>>> I fear as number III.3 makes it quite clear that usage for PBX
>>>>>> purposes is critical Ivona is not an option to be used anymore for default
>>>>>> voice prompt packages.
>>>>>>
>>>>>>
>>>>>> Br,
>>>>>> Julian
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> _________________________________________________________________________
>>>>>> Professional FreeSWITCH Consulting Services:
>>>>>> consulting at freeswitch.org
>>>>>> http://www.freeswitchsolutions.com
>>>>>>
>>>>>> 
>>>>>> 
>>>>>>
>>>>>> Official FreeSWITCH Sites
>>>>>> http://www.freeswitch.org
>>>>>> http://wiki.freeswitch.org
>>>>>> http://www.cluecon.com
>>>>>>
>>>>>> FreeSWITCH-users mailing list
>>>>>> 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
>>>>>>
>>>>>>
>>>>>
>>>>
>>>>
>>>> _________________________________________________________________________
>>>> Professional FreeSWITCH Consulting Services:
>>>> consulting at freeswitch.org
>>>> http://www.freeswitchsolutions.com
>>>>
>>>> 
>>>> 
>>>>
>>>> Official FreeSWITCH Sites
>>>> http://www.freeswitch.org
>>>> http://wiki.freeswitch.org
>>>> http://www.cluecon.com
>>>>
>>>> FreeSWITCH-users mailing list
>>>> 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
>>>>
>>>>
>>>
>>>
>>> --
>>> Michael S Collins
>>> Twitter: @mercutioviz
>>> http://www.FreeSWITCH.org
>>> http://www.ClueCon.com
>>> http://www.OSTAG.org
>>>
>>>
>>> _________________________________________________________________________
>>> Professional FreeSWITCH Consulting Services:
>>> consulting at freeswitch.org
>>> http://www.freeswitchsolutions.com
>>>
>>> 
>>> 
>>>
>>> Official FreeSWITCH Sites
>>> http://www.freeswitch.org
>>> http://wiki.freeswitch.org
>>> http://www.cluecon.com
>>>
>>> FreeSWITCH-users mailing list
>>> 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
>>>
>>>
>>
>> _________________________________________________________________________
>> Professional FreeSWITCH Consulting Services:
>> consulting at freeswitch.org
>> http://www.freeswitchsolutions.com
>>
>> 
>> 
>>
>> Official FreeSWITCH Sites
>> http://www.freeswitch.org
>> http://wiki.freeswitch.org
>> http://www.cluecon.com
>>
>> FreeSWITCH-users mailing list
>> 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
>>
>>
>
>
> --
> Michael S Collins
> Twitter: @mercutioviz
> http://www.FreeSWITCH.org
> http://www.ClueCon.com
> http://www.OSTAG.org
>
>
> _________________________________________________________________________
> Professional FreeSWITCH Consulting Services:
> consulting at freeswitch.org
> http://www.freeswitchsolutions.com
>
> 
> 
>
> Official FreeSWITCH Sites
> http://www.freeswitch.org
> http://wiki.freeswitch.org
> http://www.cluecon.com
>
> FreeSWITCH-users mailing list
> 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
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.freeswitch.org/pipermail/freeswitch-users/attachments/20130327/4eaec776/attachment-0001.html 