Feature Requests

Im extremely disapointed with the quality of this software.

How to aply colone wiath Bahasa Indonesia (Indonesian)?

This would be useful for telephony.

Some unfamiliar words are pronounced incorrectly by EVI-3 voices pronounce incorrectly, such as brand names of prescription drugs. It would be nice to have a dictionary to which developers can add entries that describe the correct pronunciation of these words using linguistic phonetic symbolic representations such as ARPAbet + CMUdict for English or IPA for International.

I’m trying to create kids story vids and I need kids voices it seems strange that a child’s voice is somehow bad it would be the content not a voice I wish you’d work on that and not restricting your value

Project generation is completely unreliable. Even simple chapters cannot be generated entirely, and multiple errors in individual paragraphs. THIS SERVICE IS THE ONLY REAON i DECIDED TO TRY HUME. It is completely useless and Ive wasted hours trying to generate a simple 6 page document. W A S T E of M O N E Y!

For version v1 of the API would be great if you increased input text limit. Now it's 5,000 chars, 11labs offer 10,000-40,000, azure tts 20,000.

Right now for POST /v0/tts .utterances[].speed` the docstring is > Speed multiplier for the synthesized speech. Extreme values below 1 and above 1.5 may sometimes cause instability to the generated output. This is a feature request to support more stability for generations with speed outside of this range.

My use case is for audiobooks, and I publish modern fantasy novels, so there are fictional names which need a custom pronunciation for. If a custom "alias" or IPA pronunciation dictionary could be configured, this would allow users like me to assign correct phonetic pronunciations for these words at one go, and have it applied globally for the project.

with synthesizeJsonStreaming There's no way to catch an error if you're out of credits. It just returns an iterable stream with nothing in it