Do you like Paul and Kate or are you fond of Mike and Crystal? Personally, I think Charles and Audrey from the UK are brilliant. This is how I started to think about synthesized voice characters after listening to them lately.
When TTS Is Useful
We can probably all agree that computer-generated voices do not have the warmth and richness of a human voice, nor can they display the range of talents found in a skilled narrator. But there are times when this option is worth exploring.
One reason to use text-to-speech (TTS) software is to provide accessibility to visually impaired persons or those who have difficulties with reading. Not all online courses are narrated and often instructions are left as text only. TTS is a way to overcome these obstacles.
TTS can also be effective as the voice of an avatar or guide. It also seems appropriate when there isn’t time or budget for recording and syncing visuals and audio, particularly for dummy or scratch audio when you need to show others how the media elements will be integrated. Finally, in an imaginative piece, TTS could be appropriate as the sound of a machine, object or computer.
Text-to-speech capabilities are integrated into Adobe Captivate, providing a way to use this feature without requiring additional software. But if your authoring tool of choice does not have this functionality, you will need to rely on external software and import the audio files. Below is a list of text-to-speech software you can explore. Listen carefully to the voices as some have demos that read back text you’ve entered. Also, many now have speakers in multiple languages.
TEXT TO SPEECH SYNTHESIZERS
iSpeech has a variety of online services and pricing models for converting TTS and downloading the files from their site. If you are developing a website, such as a learning portals, you can also connect through their API using a few lines of code and you’ve got a spoken version of your text.
NaturalSoft makes the NaturalReader software that comes in several versions, including a free one. The Professional version is most relevant to online learning as it converts files to wav/.mp3 formats and comes with two or four voices.
NeoSpeech is primarily an on-demand service, though they do license their software engine to developers. You buy credits, select a voice, enter or copy/paste text into their editor, and download the synthesized audio files.
This is an online TTS application that provides conversion of text files and documents to audio files using a variety of voice characters. There’s a free version and a paid upgrade.
This TTS software works with PowerPoint. It generates narration from PowerPoint text.
SpokenText is an online speech to text synthesizer that converts text files, documents (pdf, doc, ppt) and web pages to audio files. There are a variety of pricing structures and several voices to choose from.
NextUp sells the synthesizer TextAloud 2 for the PC and Ghostreader for the Mac. It also sells diverse character voices in different languages from the major voice publishers, including AT&T Natural Voices, Acapela Group, RealSpeak and Cepstral.
Virtual Speaker is another text to speech converter. It has a lot of options for making recordings, but potential buyers need to contact the publisher, Acapela Group, for pricing.
This is an online TTS service using Cepstral’s voices. You get access to all the character voices for one price, but this service is for online applications and devices only.
Linguatec publishes VoiceReader software, which can convert any text to audio. VoiceReader text-to-speech works in many different languages.
Just in case you’re a developer or in case your business is looking for a TTS technology, check out the Wizzard site. They produce speech applications for developers and businesses and they use AT&T’s Natural Voices.
Acapela provides diverse audio services and one of these is the creation of character voices in many different languages. Check out the variety of their voices at their site.
Cepstral is a synthetic voice publisher. You can meet Allison, Lawrence, Vittoria and many more character voices from different countries at their Demos page. Voices work with both PC and Mac.
How do you use text to speech in online learning? Tell us your experiences and recommendations.
What does it take to turn a storyboard into a script that a narrator can easily read? Whether you are recording at a studio or in-house, whether you are using a professional voice over artist or a coerced colleague, there are certain conventions that make the task easier.
Here are tips for formatting and organizing your script that apply to all types of recording—at work, home, and in professional studios. Some are known conventions and some are simply what I have discovered through trial and error.
1. Double-check for Errors
Every script has errors. It could be incorrect content or a misuse of grammar. Each error you discover during the recording session slows things down and stops the narrator’s momentum. If the error requires contacting the SME, that can be a real headache.
Therefore, go over that script with a thorough eye for detail and read it aloud. Ask someone else to review it for errors too. Making real-time corrections during a recording are not uncommon, but the less this happens, the smoother your recording will go.
2. Indicate Emphasized Words
I remember the first time I was at a recording for a script that I wrote, I was surprised that the voice talent didn’t always intonate sentences the way I intended. In hindsight, this seems obvious. How could someone else, who is not even familiar with the content, read a script with the same voice modulation I had in my mind.
That’s when I started to add emphasis in every script. Typically this is done through text formatting, such as using bold or italicized text. To avoid confusion, use one method for the entire script and communicate this convention to your narrator in a prerecording briefing.
3. Provide Pronunciation for Little Known Terms
Using terms that are unique to a field can slow down a recording. If you use medical, technical or other specialized vocabularies, find a way to communicate the pronunciation of these words and acronyms in the script. Point these out to the narrator before recording begins.
For example, you can write out the phonetic spelling of a term in brackets, so the narrator can quickly see the pronunciation. If the script uses acronyms, indicate whether the term should be pronounced by its letters or as a word. For example, when the letters alone are used, I write it with dashes, as in U-S-A.
4. Indicate Where You Need Pauses
If you allow for pauses in the script, it is easier to accommodate graphical changes on the screen, such as animations and progressive reveals. You can add an ellipsis (three dots) to the script or write the word “pause” in brackets when you need that extra half-second of silence.
Let the narrator know that at these points, you would like a pause of “one beat.” That nearly imperceptible moment of silence will help you synchronize the audio and visuals seamlessly during course production.
5. Insert Page Numbers
During a recording at a professional studio, one of my team members created a script without page numbers. The audio engineer teased him about this the whole time. That was enough to ensure I always remember page numbers. They are essential because during a live recording because everyone present will need to reference them. Also, be sure the page numbers are located in a very obvious place, such as bottom center.
6. Avoid Page Turns
When using a paper version of a script (which I find many professional voiceover artists prefer), be sure that he or she will not need to turn the page in the middle of a sentence or paragraph. The sound of paper turning usually gets picked up by the mic. Actually, this is a good tip even if the narrator is reading the script online. The time it takes to find and press the Page Down key can ruin the sound byte.
7. Name Your Audio Files
When recording for an eLearning course, I like to prepare the script so that the audio segment for each screen is associated with a unique file name. Devise a naming strategy that makes sense in your production environment. For example, m2s3.wav indicates this is the audio file you will be using in module 2-slide 3.
Therefore, my scripts have two columns—a narrow one on the left and a wider one on the right. The column on the left indicates the name of the audio file; the column on the right holds the script. This ensures that everything is well-defined for the person (even if it’s you) doing the postproduction work of breaking up the audio into smaller files.
8. Make it Easy for the Voice Talent
Regardless of whether the script is read online or from a print-out, double-space the text and use an easy to read typeface so the script is highly legible. The physical attributes of the script should be transparent to the narrating process.
Always provide the script to your narrator a few days before the recording session. Professionals always ask for a script ahead of time, so it makes sense to give it to your colleagues too. Not only will the recording have fewer retakes, your narrator will feel more comfortable and prepared.
Do you have any tips for preparing a script for production? Comment below.