Uberduck covers synthetic vocals across text to speech, voice cloning and speech-to-speech conversion, with API access for programmatic use. Its distinguishing angle is musical rather than corporate: it is built for making vocals for tracks, supporting more than 70 languages and a wide range of musical styles.
Speech-to-speech conversion is the more interesting capability. Recording a performance and converting the voice keeps the human phrasing and timing that pure text to speech flattens, which is exactly what matters when the output is music rather than narration.
Voice cloning carries obvious ethical and legal weight, and consent for any cloned voice is the user’s responsibility rather than the platform’s. For music production and creative work the toolset is strong; for straightforward narration there are more focused options.
