i’ve been experimenting with different AI voice generators lately (mainly for side projects like podcasts or narration tests), and one thing really surprised me: almost none of them let you push longer audio without hitting a hard paywall.
even the ones marketed as “free” often cap you at 30–60 seconds or ask for a credit card upfront. i get that infra costs are real, but it feels like this space might be evolving too SaaS‑first and not enough dev‑playground friendly.
curious what folks here think:
- is running high‑quality TTS/cloning really that expensive at scale, or are these just business model choices?
- do you think we’ll see truly open/free alternatives for long‑form TTS voices, or will everything drift premium like ElevenLabs?
- and if you’ve found tools that buck this trend (even research projects), i’d love to hear about them.
[link] [comments]