Automated Wave joiner.

[odomike]

Ok guys, I have this problem that have been on my neck for a long time and I ve not been able to figure out how to solve it. Neither have I come across a software that can solve it. Yea I know there are lots of wave joiner softwares out there but none of them can automate the way I want.

I need a software that can automatically insert a wave sound into an already existing wave without the help of any human. Example .... lets say there is an already existing wave file which says I live at ........................, Nigeria. What the software is meant to do is to possibly convert a text file ... for example Owerri, Imo State, and insert it into the gap in the wave file.

This is how it is supposed to work.

I have this voiced wave file which has the above words I live at .............................., Nigeria. Then I receive a text message with the following Owerri, Imo State. The job of the software is to AUTOMATICALLY convert the received text message to a wave audio format (that means that it should have Text To Speech capabilities) and also automatically insert the converted wave file into the gap in the already existing wave file and re-save with a new name. I know that this can be done manually but this is something that will need to be done hundreds of thousands over and over and no one will be able to handle such volume of conversion.

I dont know how possible this is but I need your helps guys.

Awaiting your responses

[transporterfan]

What you are looking for is a computer programmer not an off-the-shelf package.
You will need a program which can scan text files for specific strings the computer can match to a lookup table (array) of pre-defined sound clips or can pass to a speech synthesizer which can save the output to a sound file.
The second option is preferable or else you will have to have a dictionary of place names and a corresponding sound clip for every location in the entire world.
Do a search for freelance coders on Google...don't pay until it works!
Or visit the Autoit forum. One of the members there may be able throw something together if you ask nicely!

http://www.autoitscript.com/forum/

[odomike]

I have tried to go to the Autoit forum but got a 403 Forbidden page. Seems their forum doesnt like my IP. Is there any other suggestions that wouldnt involve a computer programmer?

Thanks in advance

[transporterfan]

Is there any software out there that can insert wave files into a gap in another existing wave file?

Many, but you want to automate the process "hundreds of thousands" of times...that means coding in some form or other, even if it's just a batch file. You will have to call the synthesizer with a text file, have it save the result as a sound file, join the sound files and place it somewhere that the 'player' can recognise it as a new entry and call the player and have it play it to the interested party.

There is no way around the problem other than coding. Forget it.

As for your IP, try using a proxy.

[Cornucopia]

If you are just trying to get a number of alternates filled into the gap, it makes more sense to take your original clip with the gap and make 2 clips out of it: the beforeGap clip and the afterGap clip. The gap will have to be skipped because there will be often times where the inserted Address clip will be widely varying in length, sometimes beyond the original gap length.

So you would do a batch file that would concatenate/append the files something like this:

For each {Clip} in {WatchFolder]
NewCombinedFile=beforeGap+{Clip}+afterGap
Rename NewCombinedFile.wav -> NewCombinedFile_{Clip}.wav
Next()

So then the difficult part is the text-to-speech-to-wav part.
***EDIT: Although, THIS batch text-to-wav converter might be a good place to start!!***

BTW, I DO NOT recommend using a simple batch file concatenate, as it doesn't handle the WAV headers correctly. You want a wav-aware command-line file joiner, something like shntool.

Also, for this to work seamlessly, you MUST have all the clips have the same parameters: e.g. 16bit, 44100Hz, Stereo, LPCM, LittleEndian. And it would make sense for each clip's head+tail ends to have tiny fades so the butt splice is clean with no clicks/ticks. Oh, and they should be volume- and EQ-matched, as well.

Scott

edit: If you DO go the text to speech route, to make things as seamless and professional as possible, it probably makes sense to ALSO do a text to speech recording of the beforeGap and afterGap clips. Then they would all have the same voice, and the same quality, SNR parameters, volume, etc.

[transporterfan]

If it helps, this (free version) apparently does batch conversion to WAV and can speak in different languages.

http://panopreter.com/index.php

edit: If you DO go the text to speech route, to make things as seamless and professional as possible, it probably makes sense to ALSO do a text to speech recording of the beforeGap and afterGap clips. Then they would all have the same voice, and the same quality, SNR parameters, volume, etc.

Cornucopia makes a good point there. It might even be better to concatenate the text strings, then convert it.
The resulting WAV would have the right timing and voice inflections.

[odomike]

If you are just trying to get a number of alternates filled into the gap, it makes more sense to take your original clip with the gap and make 2 clips out of it: the beforeGap clip and the afterGap clip. The gap will have to be skipped because there will be often times where the inserted Address clip will be widely varying in length, sometimes beyond the original gap length.

So you would do a batch file that would concatenate/append the files something like this:

For each {Clip} in {WatchFolder]
NewCombinedFile=beforeGap+{Clip}+afterGap
Rename NewCombinedFile.wav -> NewCombinedFile_{Clip}.wav
Next()

So then the difficult part is the text-to-speech-to-wav part.
***EDIT: Although, THIS batch text-to-wav converter might be a good place to start!!***

BTW, I DO NOT recommend using a simple batch file concatenate, as it doesn't handle the WAV headers correctly. You want a wav-aware command-line file joiner, something like shntool.

Also, for this to work seamlessly, you MUST have all the clips have the same parameters: e.g. 16bit, 44100Hz, Stereo, LPCM, LittleEndian. And it would make sense for each clip's head+tail ends to have tiny fades so the butt splice is clean with no clicks/ticks. Oh, and they should be volume- and EQ-matched, as well.

Scott

edit: If you DO go the text to speech route, to make things as seamless and professional as possible, it probably makes sense to ALSO do a text to speech recording of the beforeGap and afterGap clips. Then they would all have the same voice, and the same quality, SNR parameters, volume, etc.

Hi Cornucopia, Thanks for this awesome input. I am a lot interested in what you are explaining but I am a noob when it comes to programming. I have downloaded the trial version of textsound batch text to speech converter. I must say that I like what it has to offer but I dont know how to go about making the batch file that will automate the conversion. I do not fully understand how the shntool works, but i m still looking into it and trying to understand it further.

Can you please elaborate on this further? Like how to make the batch file? I am open to any suggestions. Anything that will make my dream come true.

I dont know, but can it be possible to create a batch file that will work with the shntool (since you recommend it over textsound batch converter) and help it to automate the conversion process?
I have already recorded a text-to-speech of the beforeGap and afterGap

Thanks to you too transporterfan for your suggestions. I am also looking into Panopreter text-to-speech converter too. It looks promising too. Thanks once again.

I am a lot happier now than before I come here. My hopes are higher that I can achieve my quest with your helps of course. Thanks once again guys.

I appreciate immensely.