Add samples/SAMPLE_INFO.md
Browse files- samples/SAMPLE_INFO.md +11 -0
samples/SAMPLE_INFO.md
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Audio samples
|
| 2 |
+
|
| 3 |
+
Short validation clips from the TranslateBlue OpenVoice V2 Core ML pipeline (2026-04). All are **16-bit PCM WAV** unless noted.
|
| 4 |
+
|
| 5 |
+
| File | Description |
|
| 6 |
+
|------|-------------|
|
| 7 |
+
| `clone_1_text_eng.wav` | English paragraph synthesized with MeloTTS (base), then **tone conversion** with Core ML E2E using a **Japanese reference** speech clip for timbre. ~5.5 s, 22050 Hz mono. |
|
| 8 |
+
| `tts_coreml_out.wav` | Output from the **Python** demo (`run_openvoice_tts_coreml.py`) using Core ML. |
|
| 9 |
+
| `swift_tts_out.wav` | Output from the **Swift** CLI (`openvoice-coreml-tts` in `SwiftOpenVoiceTTSDemo`). |
|
| 10 |
+
|
| 11 |
+
These files are for **listening tests only**; they are not training data.
|