Drop product name; pull samples pending ich-laut (espeak-ng-data) fix

README.md

@@ -17,27 +17,11 @@ tags:
 # Kokoro German "Martin" — sherpa-onnx package
 
 The German Kokoro voice **Martin**, repackaged for **[sherpa-onnx](https://github.com/k2-fsa/sherpa-onnx)**
-so it runs as **fully offline, on-device** German text-to-speech. This is the voice shipped in
-**Page Reader** (scan or paste text → read it aloud, entirely on the phone, no network at runtime).
+so it runs as **fully offline, on-device** German text-to-speech.
 
-## Samples
-
-Synthesized with *this* package via sherpa-onnx, with Page Reader's German correction lexicon
-applied (see below).
-
-**Introduction**
-
-<audio controls src="https://huggingface.co/miro-dietiker/kokoro-de-martin/resolve/main/example-intro.wav"></audio>
-
-> „Hallo, ich bin Martin, die deutsche Stimme von Page Reader. Ich lese Ihre Dokumente vor,
-> vollständig offline auf dem Gerät."
-
-**Numbers & dates** — expanded to spoken German by the app *before* synthesis:
-
-<audio controls src="https://huggingface.co/miro-dietiker/kokoro-de-martin/resolve/main/example-numbers.wav"></audio>
-
-> „Am vierundzwanzigsten Juni benötigten rund sechshundertachtzigtausend Menschen humanitäre Hilfe."
-> *(from the raw text „Am 24. Juni … 680 000 Menschen …")*
+> **Note:** audio samples are temporarily removed while a phoneme-rendering issue is fixed —
+> the bundled `espeak-ng-data` emits the German ich-laut /ç/ decomposed, so some words
+> (`ich`, `-lich`, `-ig`) don't render cleanly yet. Fix in progress.
 
 ## What's in the archive
 
@@ -46,28 +30,25 @@ applied (see below).
 - `model.onnx` — the Martin ONNX with sherpa-onnx metadata injected (`model_type=kokoro`,
   `sample_rate=24000`, `style_dim=510,1,256`, `n_speakers=1`, `id2speaker=0->martin`, …)
 - `voices.bin` — the voice style vector (510×1×256 float32)
-- `tokens.txt` and `espeak-ng-data/` — from the base k2-fsa Kokoro release
+- `tokens.txt` and `espeak-ng-data/`
 
 ## Reading German correctly (the quality layer)
 
 A good voice is only half the story — the phonemes it's given have to be right. Both Kokoro and
 Piper phonemize German with **espeak-ng**, so any word espeak's G2P gets wrong is mispronounced by
-*every* espeak-based voice. Page Reader fixes that with two reproducible, open pieces layered on
-top of this voice:
-
-- **Pronunciation-correction lexicon.** We audit espeak-ng against the **WikiPron** (Wiktionary)
-  pronunciation dictionary across ~**49k** German words, fold away espeak's *systematic*
-  conventions (which the models were trained on and render correctly), and turn the remaining
-  *word-specific* errors into a sherpa-onnx Kokoro lexicon (**2,000+ entries**) plus a corrected
-  espeak `de_dict`. Word **stress** is corrected from Wiktionary — top-5k error rate **6.8% → 1.6%**.
-  English anglicisms/names Germans actually use get Germanized pronunciations too.
+*every* espeak-based voice. Two reproducible, open pieces sit on top of this voice:
+
+- **Pronunciation-correction lexicon.** An audit of espeak-ng against the **WikiPron** (Wiktionary)
+  dictionary across ~**49k** German words folds away espeak's *systematic* conventions (which the
+  models render correctly) and turns the remaining *word-specific* errors into a sherpa-onnx Kokoro
+  lexicon (**2,000+ entries**) plus a corrected espeak `de_dict`. Word **stress** is corrected from
+  Wiktionary — top-5k error rate **6.8% → 1.6%**. English anglicisms/names get Germanized too.
 - **Language-aware text normalization.** Numbers, dates, currency and ordinals are expanded to
-  spoken German keyed to the *document's* language and surviving OCR line-wraps — e.g.
-  **„680 000" → „sechshundertachtzigtausend"** (not „sechshundertachtzig null"), „24. Juni" →
-  „vierundzwanzigsten Juni".
+  spoken German — e.g. **„680 000" → „sechshundertachtzigtausend"** (not „…achtzig null"),
+  „24. Juni" → „vierundzwanzigsten Juni" — surviving OCR line-wraps.
 
 Both are built by reproducible tooling (an espeak-vs-WikiPron audit + a Kokoro-lexicon generator,
-and this repackager) rather than hand-editing — so they can be regenerated and improved.
+and this repackager) rather than hand-editing.
 
 ## Provenance & license
 
@@ -89,4 +70,4 @@ sherpa-onnx and synthesizes a German clip to confirm it works before publishing.
 
 Extract `kokoro-de-martin.tar.bz2` and point sherpa-onnx's Kokoro config at `model.onnx`,
 `voices.bin`, `tokens.txt`, and `espeak-ng-data/` with `lang=de` (add the German lexicon for the
-corrected pronunciations shown above).
+corrected pronunciations).