The new tool arrives as data suggests that over 70% of small business owners abandon recorded video projects due to production quality concerns. Unlike many AI voice generators that struggle with cadence and emotional inflection, HeyGen Voice is engineered to replicate the specific pacing and tone of a user’s natural speech. CTO Rong Yan noted that the model was developed to ensure users sound like themselves at their best, prioritizing authentic delivery over synthetic-sounding output.

Beyond the base model, which is available for free within the platform, the company is offering a Professional Voice Clone service for $99 per month. This premium tier requires between 30 minutes and three hours of training data to capture the nuances of a specific individual. As the company scales—now reporting over $200 million in ARR and 40 million users—it continues to emphasize strict consent protocols, requiring explicit authorization before any voice or likeness is replicated through its technology.