Add TeraTTS remote-code interface

.gitattributes

@@ -1,5 +1,2 @@
 *.onnx filter=lfs diff=lfs merge=lfs -text
 *.npy filter=lfs diff=lfs merge=lfs -text
-ruaccent/dictionary/accents.json.gz filter=lfs diff=lfs merge=lfs -text
-ruaccent/dictionary/omographs.json.gz filter=lfs diff=lfs merge=lfs -text
-ruaccent/dictionary/yo_words.json.gz filter=lfs diff=lfs merge=lfs -text

README.md

@@ -12,7 +12,7 @@ tags:
 # TeraTTSv2 ONNX
 
 TeraTTSv2 is a self-contained ONNX Runtime text-to-speech release with
-selectable diffusion samplers, six voice styles, Russian stress marking, and
+selectable diffusion samplers, ten voice styles, Russian stress marking, and
 streamed audio output.
 
 This release uses the clean English/Russian 25-second teacher and its matching
@@ -24,6 +24,9 @@ eight-step CFG-3 distilled student.
 > **Important — cross-language prompts.** When an English reference voice is
 > speaking Russian, experiment with `duration_scale` below `1` (for example
 > `0.8`). It is usually a better starting point than the default `1`.
+>
+> **Recommended voices:** `ru_f1` and `ru_m5` are the preferred Russian voice
+> prompts.
 
 ## Installation
 
@@ -64,7 +67,7 @@ marking, and Unicode normalization, call `tts.normalize_text(text)`.
 
 | Control | Values | Effect |
 | --- | --- | --- |
-| `voice` | `ru_f1`, `ru_f2`, `ru_m1`, `ru_m5`, `eng_f3`, `eng_f4_whisper`, `eng_f5`, `eng_m2_whisper`, `eng_m3`, `eng_m4` | Selects a bundled precomputed voice style named after its reference audio. |
+| `voice` | `ru_f1` ★, `ru_m5` ★, `ru_f2`, `ru_m1`, `eng_f3`, `eng_f4_whisper`, `eng_f5`, `eng_m2_whisper`, `eng_m3`, `eng_m4` | Selects a bundled precomputed voice style named after its reference audio. ★ marks the recommended Russian prompts. |
 | `duration_scale` | Positive float, default `1` | Higher values produce slower, longer speech. |
 | `diffusion_model` | `distilled` (default), `teacher` | Distilled is faster; teacher supports adjustable CFG. |
 | `ruaccent_mode` | `full` (default), `dictionary` | Full uses RUAccent neural ONNX graphs plus dictionaries; dictionary mode loads dictionaries only. |
@@ -87,8 +90,12 @@ distilled sampler has CFG 3 baked into its graph.
 
 ## Language tags, numbers, and Russian stress
 
-Wrap text in `<en>…</en>` or `<ru>…</ru>`. Numbers inside those tags are
-expanded to words in the matching language before synthesis:
+Language tags are required: wrap text in `<en>…</en>` or `<ru>…</ru>`. The
+runtime rejects untagged or unbalanced input with a tag-specific error. Before
+number expansion and stress marking, it inserts spaces after punctuation and
+between a number and a following word. Characters outside the model vocabulary
+are skipped with a runtime warning. Numbers inside language tags are expanded to
+words in the matching language before synthesis:
 
 ```python
 waveform = tts.generate_speech(