| # Data and voice |
|
|
| ## Release scope |
|
|
| Inflect v2 is a fixed-voice English text-to-speech release. It is not a |
| zero-shot voice-cloning model and does not accept reference audio at inference. |
|
|
| ## Training-data summary |
|
|
| The release contains one fixed synthetic English voice. The private |
| corpus-construction and optimization recipe is not distributed. The inference |
| package does not require reference audio or call any external speech model. |
|
|
| The corpus was filtered for transcript consistency, duration, clipping, and |
| other audio-quality failures before model training. Frozen unseen-text prompts |
| were excluded from training and are included with the final evaluation |
| artifacts. |
|
|
| ## Voice status |
|
|
| The packaged voice is a synthetic model voice. This repository does not |
| redistribute a real-speaker recording corpus and does not claim that the output |
| is the identity of a real person. Users must not present generated speech as a |
| real person's recording. |
|
|
| ## Expected biases |
|
|
| - English normalization and phonemization are optimized for general North |
| American English. |
| - Proper names, uncommon locations, abbreviations, homographs, and code-switched |
| text may be mispronounced. |
| - A fixed synthetic training voice can carry narrow rhythm, prosody, spectral, |
| and pronunciation biases. |
| - The single-speaker design does not represent demographic or accent diversity. |
|
|
| ## What is not disclosed |
|
|
| The private corpus prompts, generation configuration, sampling curriculum, |
| filtering thresholds, and optimization schedule are not part of this |
| open-weight release. Their absence does not change the deployable parameter |
| count: all inference-time components and weights are included. |
|
|
