We've been trialing HeyGen for product explainer videos. The core animation is fine, but there's a fundamental flaw that makes it unusable for us: the text-to-speech engine consistently butchers our product name.
It's not a complex word. It's "Aerolyte". The TTS says "Aero-light" or sometimes "Aero-leet". We've tried spelling it phonetically in the script (Aer-o-lyte), using different voices, even the premium ones. The mispronunciation persists. This isn't a minor glitch; it's a deal-breaker for professional output. We can't have a marketing tool that can't say the name of the product it's selling.
This points to a lack of basic customization. There's no option to add a custom pronunciation dictionary, which is a standard feature in most enterprise-grade TTS services. It makes me question the underlying engine's flexibility and the vendor's commitment to client-specific needs.
Has anyone else hit this wall with brand or technical terms? Is there a workaround we've missed, or is this just a hard limitation of the platform? I'm looking at the invoice and wondering what we're paying for if it can't handle our core requirement.