VoiceGuide IVR Software Main Page
Jump to content

Text To Speech

Recommended Posts

I'm looking to use a Text to Speech program to create my .wav files.  The one I'm using is Natrual Reader.  When I put the file in and test it, it sounds like a lion roaring.  Am I doing something that isn't allowed?  When I click on the file it plays the correct content.  When I dial in it just roars.

Share this post


Link to post

Can you please .ZIP up the sound file and post the .ZIP here.

We can then see what format the sound file is in.

Also, which version of VoiceGuide are you using? If you can post latest ktTel trace from VoiceGuide's \log\ subdirectory which captures the playing of the sound file then that would let us see how the file was interpreted. (please .ZIP up the ktTel log before posting it).

 

Share this post


Link to post

The sound file attached in previous post above is in format: "PCM 48kHz 8 bit Mono"

That format is not supported.

Please see here for sound file formats supported: https://www.voiceguide.com/vghelp/source/html/soundfiles.htm

For VoiceGuide v7 the supported .WAV formats are: "PCM: 8kHz or 11kHz, 8-bit or 16-bit, Mono"

We converted the supplied file to an 8kHz version (see attached at bottom of this post), but best results would be obtained if you can get the TTS engine to directly create the .WAV file in correct format, as any conversion of format afterwards adds noise/hiss to the output sound file.

If the TTS engine is SAPI and it is installed on the VoiceGuide system then VoiceGuide would be able to use this TTS engine directly during the call.

Please see here for information on how to configure: https://www.voiceguide.com/vghelp/source/html/config_tts.htm

The generated TTS sound files are saved in VoiceGuide's \temp\ subdirectory. Those generated .WAV sound files can then be copied to another location and used in subsequent calls directly, without the need to invoke TTS engine again. (ie. just play it as any other .WAV file)

 

MainMessage_pcm8kHz.wav

Share this post


Link to post

I have same issue. Can you recommend speech Syntesizer program? I tried textsound but client is not happy they say the voice is bad. Any idea why?
We talk about wav files.

I need command line interface syntesizer

Share this post


Link to post

Many 3rd party programs will just use whatever TTS engines are installed on your computer to generate the sound files.

Suggest you take a look at and demo this one:

https://www.cepstral.com/en/telephony/windows

There are a number of other providers of TTS engines as well.

If you install the TTS engine on VoiceGuide server then instruction on configuring VoiceGuide to use that TTS engine are here:

https://www.voiceguide.com/vghelp/source/html/config_tts.htm

 

Soon VoiceGuide will also support playing of Google Cloud TTS. You can demo it here: https://cloud.google.com/text-to-speech/

Share this post


Link to post

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×