Lab-MSP
/

ChiReSSD

@@ -44,8 +44,8 @@ Fine-tuned from the StyleTTS2 LibriTTS second-stage checkpoint by
 here** — obtain it from the upstream release. The frozen helper models (ASR text aligner, JDCNet
 pitch extractor, PL-BERT) likewise ship inside the upstream repository and are not redistributed.
-ChiReSSD's modifications to StyleTTS2 is published as patch files in the code
-repository.
 ## Intended use
@@ -54,13 +54,16 @@ disorders.
 ## Out of scope
-- Not a medical device. No diagnostic or treatment decision should rest on its output.
 - Not a replacement for assessment by a licensed speech-language pathologist.
 ## Training data
 Child speech from UltraSuite/CLP-derived recordings, obtained under their own data use
-agreements. The corpus is not released here and is not redistributable: it is identifiable
 child clinical speech. See `DATA.md` in the code repository for how to obtain comparable data;
 <https://huggingface.co/datasets/changelinglab/ultrasuite-benchmark> is a suitable starting point
 for fine-tuning your own model.
@@ -99,10 +102,24 @@ articulation itself.
 `alpha` is high on purpose. A low `alpha` leans on the reference acoustics, which is exactly
 where the disordered articulation lives.
-Synthesis is stochastic. The initial style latent is zeros rather than a Gaussian draw, but
 the ADPM2 sampler is ancestral and adds fresh noise at every step, so repeated calls differ in
 waveform and in duration. Pass a `seed` for reproducible output.
 ## Usage
 ```python
@@ -119,7 +136,22 @@ wav = model.synthesize(
 Run `chiressd-setup` first: it clones and patches the upstream StyleTTS2 checkout that supplies
 the architecture and frozen helper models.
 ## Citation
-Karen Rosero, Eunjung Yeo, David R. Mortensen, Cortney Van’t Slot, Rami R. Hallac, and Carlos Busso "Generative Reconstruction of Pediatric
-Disordered Speech for Automated Clinical Evaluation," in Proc. of the 2026 IEEE Spoken Language Technology Workshop, December, 2026.

 here** — obtain it from the upstream release. The frozen helper models (ASR text aligner, JDCNet
 pitch extractor, PL-BERT) likewise ship inside the upstream repository and are not redistributed.
+ChiReSSD's own modification to StyleTTS2 is two lines, published as patch files in the code
+repository rather than as a fork.
 ## Intended use
 ## Out of scope
+- **Not a medical device.** No diagnostic or treatment decision should rest on its output.
 - Not a replacement for assessment by a licensed speech-language pathologist.
+- Not validated for populations outside the training corpus — it was adapted to children aged
+  5–12 with a Central Scottish accent, and the phonemizer default (`en-gb`) reflects that.
+- Not a general-purpose voice cloning system.
 ## Training data
 Child speech from UltraSuite/CLP-derived recordings, obtained under their own data use
+agreements. **The corpus is not released here and is not redistributable**: it is identifiable
 child clinical speech. See `DATA.md` in the code repository for how to obtain comparable data;
 <https://huggingface.co/datasets/changelinglab/ultrasuite-benchmark> is a suitable starting point
 for fine-tuning your own model.
 `alpha` is high on purpose. A low `alpha` leans on the reference acoustics, which is exactly
 where the disordered articulation lives.
+**Synthesis is stochastic.** The initial style latent is zeros rather than a Gaussian draw, but
 the ADPM2 sampler is ancestral and adds fresh noise at every step, so repeated calls differ in
 waveform and in duration. Pass a `seed` for reproducible output.
+## Evaluation
+See the paper. **This model repository does not reproduce the paper's reported numbers**, and the
+code repository does not either: the evaluation corpora are not redistributable, so the numbers
+cannot be independently recomputed from what is published here.
+## Ethical considerations
+This model is trained on identifiable clinical recordings of children and is designed to
+reproduce a child's vocal identity accurately. High speaker similarity is the method's goal and
+therefore also its misuse vector: the same property that makes feedback usable in a child's own
+voice makes the model a voice-cloning tool for a minor. Use it only with appropriate consent and
+ethical oversight.
 ## Usage
 ```python
 Run `chiressd-setup` first: it clones and patches the upstream StyleTTS2 checkout that supplies
 the architecture and frozen helper models.
+## Checkpoint provenance
+Derived from training checkpoint `epoch_2nd_00003.pth` of the v5 run by keeping `state['net']`
+only, removing the `module.` prefix that DataParallel added to ten of the thirteen submodules,
+and making every tensor detached, CPU-resident and contiguous. Precision is unchanged (float32;
+no fp16 cast, which would alter outputs).
+- 2258 MB → 767 MB; the 1491 MB removed is optimizer state, which inference never reads
+- `sha256` of `model.pth`: `6c5faf5b4967b4d26cb4c58d4517dcc477729a579039316741bd88b9bd0f16e0`
+- Verified bit-exact against the training checkpoint it was stripped from: under a fixed seed,
+  the two produce an identical 256-d style vector and identical waveform samples, so the strip
+  provably changes no output
+Full details in `manifest.json`.
 ## Citation
+Rosero, Yeo, Mortensen, Van't Slot, Hallac, and Busso. *Generative Reconstruction of Pediatric
+Disordered Speech for Automated Clinical Evaluation.*