The Text to Speech module explained (TTS)

In this article, we want to have a closer look at the Text to speech module (TTS).

The TTS module works similar to an audio player but is much more flexible. It can be used to read out a text to the customer. The text does not have to be uploaded but can be inserted directly into the module.

Let's go through the settings one by one:

In the upper part of the module you can choose the language of the text you are entering into the module. babelforce offers 70+ languages out of the box. Please note that we cannot translate voice input out of the box. So, if you choose "English, Great Britain" as the language for instance, the module reads out the text with an English accent (even if it might be French ;)).

mceclip3.png

Additionally to the language, you can choose the gender of the reading voice. You can choose between neutral, female and male.

Please note that the gender is only applied if you choose one of the standard voices: babelforce offers 9 voices out of the box. The WaveNet voices are pregonfigured by google, so you can not change their gender. Just try out which voice fits best for your language.

Now, let's look at the field to configure your text:

mceclip1.png

In this field you can enter the text you want to have read out to the customer. There are many different possibilities to tweak your text to sound natural. However, the only adjustment needed for the text to be read out are the brackets <speak> at the beginning and <break strength="x-strong"/></speak> at the end of the text. You can add them by clicking on the symbol "</>".

To tweak and customize your text, babelforce offers the Speech Synthesis Markup Language (SSML). For a detailed description of the features, please have a look at this article by Google.

Let's have a quick look on all of them:

mceclip0.png Click on this button to have your configured text read to you (not active yet).
mceclip1.png Insert pauses into your text, e.g. "Hello, <break strength="weak"/>how are you?"
mceclip2.png Emphasize words or parts of a word by marking the part you want to emphasize and clicking on the symbol.
mceclip3.png Here you can choose the language of the text you are configuring. You can disregard this setting since you already configured the language above. 
mceclip4.png With this element, you can add paragraphs to your text.
mceclip7.png This element lets you indicate information about the type of text construct that is contained within the element. It also helps specify the level of detail for rendering the contained text.
mceclip6.png Here you can configure text to be interpreted as a date.
mceclip8.png With this setting you can substitue text. Let's say you entered a placeholder into the TTS module, but the placeholder has not been filled with data. You can then enter a fallback text for the placeholder.
mceclip9.png Use this to use the phonetic alphabet.
mceclip10.png As explained above, set your text into the speak brackets to be read in your call flow.
mceclip12.png Use this symbol to validate if your Text has been set up correctly.

 

 

Once you configured your text, you need to choose an after-flow of your TTS module:

mceclip2.png

 

Have more questions? Submit a request