>>105804805
The text to speech application Openaudio S1 Mini can produce 96 second audio files. Plus it has emotion tags like (joyful) and (sad).

Link for tags
https://huggingface.co/fishaudio/openaudio-s1-mini
Link for local app
https://huggingface.co/spaces/fishaudio/openaudio-s1-mini/tree/main

Sample:
https://vocaroo.com/1boIKhWykbuP