Publish adaptation toolkit documentation

README.md

@@ -41,6 +41,10 @@ Fixed-voice English TTS with deterministic seeds, long-text handling, and CPU or
 
 <p align="center"><strong>9,356,513 deployable parameters</strong> · <strong>37.53 MB FP32</strong> · <strong>24 kHz mono output</strong></p>
 
+> **New: public adaptation toolkit**
+>
+> Prepare data, audit train/validation splits, adapt a fixed voice or language, resume training, evaluate checkpoints, and export PyTorch or ONNX packages with the [Inflect adaptation toolkit](https://github.com/owenawsong/Inflect/tree/main/finetune). Adapted quality is experimental and depends on the dataset, frontend, and fluent-speaker evaluation.
+
 ---
 
 Inflect v2 uses one public API across two sizes: **Micro prioritizes quality below 10M parameters; Nano prioritizes footprint below 4M.**
@@ -54,6 +58,7 @@ Inflect v2 uses one public API across two sizes: **Micro prioritizes quality bel
 | [Evaluation](#evaluation) | [Controls and long text](#controls-determinism-and-long-text) |
 | [Choose Micro or Nano](#choose-the-right-inflect) | [Data and adaptation](#data-voice-and-adaptation-status) |
 | [Run locally](#run-locally) | [Exports and quantization](https://huggingface.co/owensong/Inflect-Micro-v2/blob/main/docs/EXPORTS.md) |
+| [Adapt a voice or language](https://github.com/owenawsong/Inflect/tree/main/finetune) | [Training and export workflow](https://github.com/owenawsong/Inflect/tree/main/finetune/docs/TRAINING.md) |
 | [Package map](#package-map) | [Evaluation and raw protocol](https://huggingface.co/owensong/Inflect-Micro-v2/blob/main/docs/EVALUATION.md) |
 | [Limitations](#limitations) | [Deployment guide](https://huggingface.co/owensong/Inflect-Micro-v2/blob/main/docs/DEPLOYMENT.md) |
 
@@ -332,7 +337,7 @@ Long passages are punctuation-aware chunks, not one unlimited autoregressive pas
 
 The release contains one fixed synthetic English voice. The package does not redistribute a real-speaker recording corpus, does not claim the voice as the identity of a real person, and requires no reference audio or external model at inference.
 
-This release is inference-first. New-voice and new-language adaptation are **not currently validated or supported**. A new voice would replace the fixed speaker rather than add a selectable speaker; language adaptation also requires rebuilding normalization, phonemes, symbols, embeddings, and training data. See [`docs/DATA_AND_VOICE.md`](https://huggingface.co/owensong/Inflect-Micro-v2/blob/main/docs/DATA_AND_VOICE.md) and [`docs/FINETUNING.md`](https://huggingface.co/owensong/Inflect-Micro-v2/blob/main/docs/FINETUNING.md).
+The base release remains inference-first, but a public **experimental fixed-voice and language adaptation workflow** is now available. A new voice replaces the built-in speaker rather than adding runtime voice cloning. A new language requires owned or licensed speech data, a compatible phoneme frontend, symbol migration, retraining, and fluent-speaker evaluation. Start with the [Inflect adaptation toolkit](https://github.com/owenawsong/Inflect/tree/main/finetune), then review [`docs/DATA_AND_VOICE.md`](https://huggingface.co/owensong/Inflect-Micro-v2/blob/main/docs/DATA_AND_VOICE.md).
 
 </details>