All models
experimental

WhisperSpeech

MIT

Whisper, but inverted — TTS by 'unwrapping' OpenAI's ASR

WhisperSpeech, from Collabora, flips OpenAI's Whisper architecture around to generate speech instead of transcribing it. It's trained entirely on properly licensed open data — a deliberately transparent, commercially safe stack — with voice cloning from reference audio.

Voices

1 voice
Neutral
Defaultdefault
—
NeutralEmotionalNumbers & Dates

The scripts

Neutral
“The quick brown fox jumps over the lazy dog near the riverbank.”
Emotional
“I can't believe you actually did it. This is incredible!”
Numbers & Dates
“On March 14th, 2025, the team raised $4.2 million at a 38% margin.”

Install

pip install whisperspeech

WhisperSpeech FAQ

Do I need a GPU to run WhisperSpeech?

For practical speeds, yes — plan on roughly 6 GB of VRAM. If you'd rather not manage a GPU, hosted access to every model in this directory is coming via OpenSpeech Cloud.

Can WhisperSpeech clone voices?

Yes — WhisperSpeech supports voice cloning from reference audio.

What license is WhisperSpeech released under?

WhisperSpeech is released under the MIT license. Always check the repository for the exact terms — some models license code and weights separately.

What languages does WhisperSpeech support?

WhisperSpeech supports English, Polish.

Is there a hosted WhisperSpeech API?

Yes — OpenSpeech Cloud is launching hosted, pay-per-minute access to WhisperSpeech and every other model in this directory. No CUDA, Docker, or GPU needed.