I think Audiosonic is the best human sounding text to speech platform I've heard. I listened to all of the English voice examples, but didn't see a reference to other languages that I could also demo. Regardless, since English is the primary language used in my business, I'd like to first congratulate you on this remarkable offering, and then offer a couple of comments/suggestions if I may:
1. One very obvious thing I noticed about "Michael" in particular, but also "Matilda" was the very human, but distracting plosive sounds that are made when speaking too closely into a microphone. Very interesting that this should be on an AI voice, but even so, it's an unwanted artifact in an otherwise very high quality voice.
2. Like previous reviewers, I would also like to see a method to adjust parameters such as (in my words): attitude, speed, timbre, pitch, and tone. There may be others, as well, such as "musicality" or the quality of varying voice pitch, speed, and emphasis throughout a spoken phrase or sentence, apart from the "normal" tone established.
3. Cloning of one's personal voice, with the addition of the parameters mentioned above, could truly be life-changing for some of us with less-than-perfect voices for audio presentations.
That is all I can come up with at present, but I am very excited about your new product offering. Even without the improvements mentioned, it could quickly become a very useful tool for me in YT videos, podcasts, and audio clips attached to product offerings.
Report
18 views