Farewell, HAL 9000; Hooray for Google TTS
I see this HAL 9000 indicator every time I convert a simulation to an audio project using the Windows Text-To-Speech (TTS) engine. Why should we keep using it in 2023 when a super cool alternative is offered by Google Cloud Platform and natively supported by SAP Enable Now? Yes, this blog post will be about setting up the Google TTS in the SAP Enable Now, so you could forget about robotic Dave and insensitive Zira.
Google Cloud Platform Setup
- You must create a Google Cloud Platform (GCP) account. The trial one is perfectly fine 🙂 You can use this link to register. This Quick Start video from Google is also very helpful.
- Once the registration is completed and you log into the GCP, we must create a new project. Expand the hamburger menu, find the IAM & Admin – Create a Project. Otherwise, try this direct link.
- Assign a name to your new project. I’ll go with sen-tts.
- After creating the project, navigate to APIs & Services, choose Enable APIs and Services
- Look for “Text to Speech” using the search bar.
- Choose this one and enable it
- How much does it cost? The official GCP pricing you can get here. For my proudly Canadian friends and customers, converting up to 4 million characters to speech will cost 0 (zero) dollars. For every additional 1.000.000 characters, we need to pay just 5.31 dollars. Canadian, of course. In USD, it’s even cheaper. The 4M free characters quota is provided every month. So, probably, for many of us, this service will be absolutely free.
- The GCP will automatically redirect you to the newly activated TTS Service Page, go back to the APIs & Services, and select Credentials to create an API key.
- Save your API key. Also, it might be a bright idea to limit the usage of this API key to the Google TTS service only. You can do it by clicking the Edit Key link on the newly generated API key popup or by clicking on the API key on the APIs and Services – Credentials screen
Congrats! You’ve completed the setup on the GCP side. Now, let’s move to the SAP Enable Now Producer.
SAP Enable Now Producer Setup
- Open the simulation project.
- Convert it to an audio project (if it has no audio) by clicking Project – Convert to Audio Project
- Select Audio – Generate Text-to-Speech Audio
- Then click the Configure Text-to-Speech Services
- Select the Google Text-to-Speech, paste your API key and click Validate.
Now enjoy more than 40 variations of English voice, 15+ options for each content language supported by SAP Enable Now.
The last thing: you can make the Google TTS sound totally natural i.e. by telling it where to make a long pause, what words to emphasize, how to pronounce particular acronyms, numbers etc, by adding the Speech Synthesis Markup Language.
P.S. You can also enable this feature for all content developers in the Authoring settings – Audio