What HeyGen shipped
HeyGen introduces Voice Speech-to-Speech, a feature designed to capture the exact emotion, pacing, and emphasis of a user's delivery. By replacing text-to-speech prompting with direct audio input, it maps a casual phone or laptop recording onto a studio-quality AI voice.
The update targets the limitations of text prompting, allowing users to retain control over nuances like whispers, laughs, and accents while upgrading the fidelity of the audio itself.
How the motion works
When demonstrating an audio-first feature like speech-to-speech, the edit should prioritize the sound design. A motion designer might rely on a hard-cut to switch between the raw input recording and the studio-quality output. This lets the contrast in audio fidelity carry the moment, stripping away complex visual transitions that might distract from the core product claim.
To emphasize the nuances of delivery—whispers, laughs, and accents—kinetic-typography serves as an effective tool to visually represent pacing and emphasis. Scaling up specific words exactly as they are stressed in the audio track grounds the abstract concept of human emotion in a precise visual system.
Visualizing the transformation from a casual phone recording to a polished AI voice requires clear anchors. A match-cut between a lo-fi UI element representing the phone input and a clean, high-resolution interface for the HeyGen Voice output creates an immediate understanding of the upgrade. Adding a subtle punch-in on the final studio-quality result reinforces the focus on the end product.
What to steal from it
- Let the audio dictate the visual pacing when demonstrating voice or sound-based products.
- Use kinetic typography to make abstract concepts like vocal emphasis and pacing visible on screen.
- Contrast low-fidelity input with high-fidelity output using direct, unembellished cuts to highlight the transformation.
- Anchor feature announcements in the specific user action required, such as recording directly on a phone or laptop.
Want a video like this for your product?
Impractical cuts one from a single prompt — the same motion craft, in about twenty minutes.
