Feature #153

Provide some pre created TTS for standing speech elements during install

Added by tshif over 2 years ago. Updated over 1 year ago.

Status:Rejected Start:06/24/2009
Priority:Normal Due date:
Assigned to:- % Done:

0%

Category:-
Target version:-

Description

The download of the report and the translation of the file from text to speech takes some time.

Proposal: During the install process of the module, create with flite or swift the different announcements (your report has been downloaded, etc....). Then you can:

- for the initial zip code request: use background() to allow to enter the zip code before the end of the announcement.
- launch the translation before playing the other announcements to save some time.
(by patrick_elx)


Related issues

related to Weather by ZIP Code - Feature #154: Allow user to enter zip code during prompt playback Feedback 06/24/2009

History

Updated by tshif over 2 years ago

Ok - I see a lot of potential for failure in trying to do this. If we cant find flite or swift for example. Also - Im not sure just how much time we would actually save. I know it would conserve licenses (cepstral) - but well still be waiting just as long for the report to be parsed. Thats where the real time is spent.

This case is desinged to foster discussion about this feature request. Please comment freely.

Updated by patrick_elx over 2 years ago

You will save time if you start the TTS for the report, then play the 'we are downloading your report now" or whatever other sentence you want to say. I'm not sure if it's possible that way as we are waiting for the answer of the engine before doing something else.

On the other hand it should be possible to send the playback of the message to another non blocking thread, then start the TTS of the report immediately, while the playback is still playing. We'll save a few seconds there.

Updated by tshif over 2 years ago

We can not prepackage custom cepstral created voices created using the licenses we use for pbx work. Plus, they might not be using allison voice. They could be using the David voice.

Could we create the sound files using the users installed flite and cepstral (if present) upon install of the module? If we did it that way, the user could rerun the module installer if they add cepstral and need the new sound files created.

We would have to have a bullet proof way of making sure this got done, or the entire module would fail to operate.

Updated by Bitnetix over 2 years ago

I don't like where this is going, to be honest. We're talking about making assumptions about voices, checking to see what's installed, assuming it's installed where we think it is, and so forth. User might have two (or more) voices installed and use one for incoming calls and one for outgoing messages, or whatever. Making this into a menu system for configuration would at least allow that to happen, but I think it's a lot of work for little gain.

I think that the few seconds of waiting on the user's behalf also adds to the appearance that "real work is being done." And makes it easier to justify not doing the work in this ticket. :-)

My opinion: reject.

Updated by tshif over 2 years ago

Patrick - do you have any further thoughts to encourage adoption of this idea?

Updated by tshif over 2 years ago

  • Target version deleted (Weather by ZIP Code - Future Releases)

Updated by tshif over 1 year ago

  • Status changed from Feedback to Rejected

Not much demand for this, and Bitnetix is correct - wed be making a LOT of assumtions about the target system. We are safer keeping it the way it is. The pauses are not THAT bad.

Also available in: Atom PDF