The Lone Voice in the Wilderness

Once upon a time there was a place online where you could animate ready-made 3D characters in a scene and have them speak. The program was called XtraNormal and was found at http://xtranormal.com/. It was a lot of fun to play with, but sadly it was suspended for reasons unknown to me.

One its best features was the myriad variety of different voices available for the digital actors. Some of the voices had dialects that were clearly identifiable with different cultures. There were dozens of voices in different dialects and languages. But it's no longer available, at least for the time-being.

In my quest for similar products I found Plotagon and MovieStorm.

Plotagon has two basic male and female voices that can pronounce most words clearly, but vowels in certain words are shortened so much they confuse the listener. The scenes are limited to no more than two characters. The usual work-around to have more than two characters in a scene must be done in post-production with two separate scenes using the same set, using software like Sony Movie Studio or Movie Maker which comes with most PC's

MovieStorm on the other hand allows for as many characters in a scene as your computer's memory will allow. You can build multiple stages for the same scene and have your main camera jump from one scene to the next, jump between characters for dialogue, or have the camera truck, pan, tilt, and zoom around the scene.

On top of that, you can add multiple cameras, but each camera records the entire scene from its perspective, so when you get to the editing part, you have several minutes of extra footage from multiple cameras that will need to be cut out. In the orignal XtraNormal, you could have multiple cameras, but they recorded the scene as if a live technical director was switching between them during the recording process, so there was no excess footage.

MovieStorm allows you to use your microphone to record the dialogue, which means you can have your friends read their lines from a script and email you the voice file. However, if you wake up in the middle of the night with an inspired scene you want to crank out before dawn, MovieStorm has only one Text-To-Speech (TTS) voice, and it's very monotone.

In my quest for a solution, I tested TTS software called Balabolka which will convert a document or text to an audio file, but it only utilizes the two voices that came with my computer, and if I wanted to have it read a script, it requires breaking up each actor's lines into separate files. Which is fine because MovieStorm requires each actor's line to be a separate file anyway. It's moderately tedious.

At one point I created a scene using Plotagon, created a YouTube video, then with Free YouTube Downloader I captured the audio file and used it in MovieStorm. Still, it was just two voices, one male, one female.

Then I found Verbose from NCH software. Another TTS application that allows you to change the pitch, speed and volume of the male and female voice installed on my system (in my case there are only two and I have Windows 10). However, NCH also has available a program called Voxal Voice Changer, which can alter your own voice, or an existing audio file, to sound like many different characters, and comes pre-loaded with settings for such voices as "Frankenstein," "Pirate" and even "Old Female" 

This is all great, but it's not what I really want.

What I want is a TTS program that can read a script with more voices available to the actors. One which can interpret emphasis descriptors within the dialogue like "[angry]" or "[sad]". One which will clip the lines and automatically save them as separate audio files for me.