
I like Fish Audio for its high-quality cloned voices and the incredible speed it provides. The ability to create almost infinite cloned voices using public voices is quite impressive. I love the expressions you can use and all the effects that come with it. It's basically the best audio voice generation I've tested, and I've tested quite a lot. Additionally, it comes with an API and you can run it locally if you have a strong enough graphics card, which I find bloody amazing. The setup is pretty effortless as well; the API is self-explanatory and I didn't face any issues integrating it. Even running it locally was straightforward. And most important it cost 1/10 the price of competitors. Review collected by and hosted on G2.com.
I'm missing a feature in Fish Audio that allows combining two(or multiple) speakers to make one output automatically, something I could do in Gemini TTS to create dialogues. I have to solve it manually, which is less convenient. Review collected by and hosted on G2.com.