voicefest

_script/voicefest.py

@@ -0,0 +1,160 @@
+#!/usr/bin/env python3
+import hashlib
+import json
+import unittest
+from pathlib import Path
+
+_DIR = Path(__file__).parent
+_REPO_DIR = _DIR.parent
+
+
+class VoiceTest(unittest.TestCase):
+    def test_voices(self):
+        for onnx_path in _REPO_DIR.rglob("*.onnx"):
+            with self.subTest(onnx_path=onnx_path):
+                self.assertGreater(onnx_path.stat().st_size, 0, "Empty onnx file")
+
+                # Load JSON config for voice
+                config_path = onnx_path.parent / f"{onnx_path.name}.json"
+                with open(config_path, "r", encoding="utf-8") as config_file:
+                    config = json.load(config_file)
+
+                # Verify config
+                self.assertIn(
+                    "piper_version", config, "Missing piper_version in config"
+                )
+                self.assertIn("language", config, "Missing language in config")
+                self.assertIn("dataset", config, "Missing dataset in config")
+                self.assertIn(
+                    "quality", config["audio"], "Missing audio quality in config"
+                )
+
+                # Verify directory structure
+                # <lang_family>/<lang_code>/<dataset>/<quality>/
+                quality_dir = onnx_path.parent
+                dataset_dir = quality_dir.parent
+                lang_code_dir = dataset_dir.parent
+                lang_family_dir = lang_code_dir.parent
+
+                self.assertEqual(
+                    lang_family_dir.name,
+                    config["language"]["family"],
+                    "Wrong lang family dir",
+                )
+                self.assertEqual(
+                    lang_code_dir.name,
+                    config["language"]["code"],
+                    "Wrong lang code dir",
+                )
+                self.assertEqual(
+                    dataset_dir.name, config["dataset"], "Wrong dataset dir"
+                )
+                self.assertEqual(
+                    quality_dir.name, config["audio"]["quality"], "Wrong quality dir"
+                )
+
+                # Verify file names
+                file_lang_code, file_dataset, file_quality = onnx_path.stem.split("-")
+                file_lang_family = file_lang_code.split("_", maxsplit=1)[0]
+
+                self.assertEqual(
+                    file_lang_family,
+                    config["language"]["family"],
+                    "Wrong lang family file",
+                )
+                self.assertEqual(
+                    file_lang_code, config["language"]["code"], "Wrong lang code file"
+                )
+                self.assertEqual(file_dataset, config["dataset"], "Wrong dataset file")
+                self.assertEqual(
+                    file_quality, config["audio"]["quality"], "Wrong quality file"
+                )
+
+
+def run_tests():
+    runner = unittest.TextTestRunner()
+    runner.run(unittest.makeSuite(VoiceTest))
+
+
+# -----------------------------------------------------------------------------
+
+
+def write_voices_json():
+    # {
+    #   "<family>_<region>-<dataset>-<quality>": {
+    #     "name": "<dataset>",
+    #     "language": "<family>_<region>",
+    #     "quality": "<quality>",  // x_low, low, medium, high
+    #     "num_speakers": int,
+    #     "speaker_id_map": {
+    #       "name": int,
+    #       ...
+    #     }
+    #     "files": {
+    #       "relative/path/to/file": {
+    #         "size_bytes": int,
+    #         "md5_digest": str,        // hex
+    #       },
+    #       ...
+    #     }
+    #   },
+    #   ...
+    # }
+    voices = {}
+
+    for onnx_path in _REPO_DIR.rglob("*.onnx"):
+        voice_dir = onnx_path.parent
+        config_path = voice_dir / f"{onnx_path.name}.json"
+        with open(config_path, "r", encoding="utf-8") as config_file:
+            config = json.load(config_file)
+
+        quality = config["audio"]["quality"]
+        dataset = config["dataset"]
+        lang_code = config["language"]["code"]
+        voice_key = f"{lang_code}-{dataset}-{quality}"
+
+        model_card_path = voice_dir / "MODEL_CARD"
+        assert model_card_path.exists(), f"Missing {model_card_path}"
+
+        voices[voice_key] = {
+            "name": dataset,
+            "language": lang_code,
+            "quality": quality,
+            "num_speakers": config["num_speakers"],
+            "speaker_id_map": config.get("speaker_id_map", {}),
+            "files": {
+                str(file_path.relative_to(_REPO_DIR)): {
+                    "size_bytes": file_path.stat().st_size,
+                    "md5_digest": get_file_hash(file_path),
+                }
+                for file_path in (
+                    onnx_path,
+                    config_path,
+                    model_card_path,
+                )
+            },
+        }
+
+    with open(_REPO_DIR / "voices.json", "w", encoding="utf-8") as voices_file:
+        json.dump(voices, voices_file, indent=4, ensure_ascii=False)
+
+
+def get_file_hash(path, bytes_per_chunk: int = 8192) -> str:
+    """Hash a file in chunks using md5."""
+    path_hash = hashlib.md5()
+    with open(path, "rb") as path_file:
+        chunk = path_file.read(bytes_per_chunk)
+        while chunk:
+            path_hash.update(chunk)
+            chunk = path_file.read(bytes_per_chunk)
+
+    return path_hash.hexdigest()
+
+
+# -----------------------------------------------------------------------------
+
+if __name__ == "__main__":
+    run_tests()
+
+    print("Writing voices.json")
+    write_voices_json()

ca/ca_ES/upc_ona/medium/ca_ES-upc_ona-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "ca"
@@ -479,5 +480,11 @@
     "num_symbols": 256,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "ca_ES",
+        "family": "ca",
+        "region": "ES"
+    },
+    "dataset": "upc_ona"
 }

ca/ca_ES/upc_ona/x_low/ca_ES-upc_ona-x_low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "x_low"
     },
     "espeak": {
         "voice": "ca"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "ca_ES",
+        "family": "ca",
+        "region": "ES"
+    },
+    "dataset": "upc_ona"
 }

ca/ca_ES/upc_pau/x_low/ca_ES-upc_pau-x_low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "x_low"
     },
     "espeak": {
         "voice": "ca"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "ca_ES",
+        "family": "ca",
+        "region": "ES"
+    },
+    "dataset": "upc_pau"
 }

da/da_DK/talesyntese/medium/da_DK-talesyntese-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "da"
@@ -479,5 +480,11 @@
     "num_symbols": 256,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "da_DK",
+        "family": "da",
+        "region": "DK"
+    },
+    "dataset": "talesyntese"
 }

de/de_DE/eva_k/x_low/de_DE-eva_k-x_low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "x_low"
     },
     "espeak": {
         "voice": "de"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "de_DE",
+        "family": "de",
+        "region": "DE"
+    },
+    "dataset": "eva_k"
 }

de/de_DE/karlsson/low/de_DE-karlsson-low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "low"
     },
     "espeak": {
         "voice": "de"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "de_DE",
+        "family": "de",
+        "region": "DE"
+    },
+    "dataset": "karlsson"
 }

de/de_DE/kerstin/low/de_DE-kerstin-low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "low"
     },
     "espeak": {
         "voice": "de"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "de_DE",
+        "family": "de",
+        "region": "DE"
+    },
+    "dataset": "kerstin"
 }

de/de_DE/pavoque/low/de_DE-pavoque-low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "low"
     },
     "espeak": {
         "voice": "de"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "de_DE",
+        "family": "de",
+        "region": "DE"
+    },
+    "dataset": "pavoque"
 }

de/de_DE/ramona/low/de_DE-ramona-low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "low"
     },
     "espeak": {
         "voice": "de"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "de_DE",
+        "family": "de",
+        "region": "DE"
+    },
+    "dataset": "ramona"
 }

de/de_DE/thorsten/low/de_DE-thorsten-low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "low"
     },
     "espeak": {
         "voice": "de"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "de_DE",
+        "family": "de",
+        "region": "DE"
+    },
+    "dataset": "thorsten"
 }

de/de_DE/thorsten/medium/de_DE-thorsten-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "de"
@@ -473,5 +474,11 @@
     "num_symbols": 256,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "de_DE",
+        "family": "de",
+        "region": "DE"
+    },
+    "dataset": "thorsten"
 }

el/el_GR/rapunzelina/low/el_GR-rapunzelina-low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "low"
     },
     "espeak": {
         "voice": "el"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "el_GR",
+        "family": "el",
+        "region": "GR"
+    },
+    "dataset": "rapunzelina"
 }

en/en_GB/alan/low/en_GB-alan-low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "low"
     },
     "espeak": {
         "voice": "en-gb-x-rp"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "en_GB",
+        "family": "en",
+        "region": "GB"
+    },
+    "dataset": "alan"
 }

en/en_GB/alan/medium/en_GB-alan-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "en-gb-x-rp"
@@ -479,5 +480,11 @@
     "num_symbols": 256,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "en_GB",
+        "family": "en",
+        "region": "GB"
+    },
+    "dataset": "alan"
 }

en/en_GB/alba/medium/en_GB-alba-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "en-gb-x-rp"
@@ -479,5 +480,11 @@
     "num_symbols": 256,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "en_GB",
+        "family": "en",
+        "region": "GB"
+    },
+    "dataset": "alba"
 }

en/en_GB/aru/medium/en_GB-aru-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "en-gb-x-rp"
@@ -492,5 +493,11 @@
         "07": 10,
         "04": 11
     },
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "en_GB",
+        "family": "en",
+        "region": "GB"
+    },
+    "dataset": "aru"
 }

en/en_GB/jenny_dioco/medium/en_GB-jenny_dioco-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "en-gb-x-rp"
@@ -479,5 +480,11 @@
     "num_symbols": 256,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "en_GB",
+        "family": "en",
+        "region": "GB"
+    },
+    "dataset": "jenny_dioco"
 }

en/en_GB/northern_english_male/medium/en_GB-northern_english_male-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "en-gb-x-rp"
@@ -473,5 +474,11 @@
     "num_symbols": 256,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "en_GB",
+        "family": "en",
+        "region": "GB"
+    },
+    "dataset": "northern_english_male"
 }

en/en_GB/southern_english_female/low/en_GB-southern_english_female-low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "low"
     },
     "espeak": {
         "voice": "en-gb-x-rp"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "en_GB",
+        "family": "en",
+        "region": "GB"
+    },
+    "dataset": "southern_english_female"
 }

en/en_GB/vctk/medium/en_GB-vctk-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "en-gb-x-rp"
@@ -589,5 +590,11 @@
         "p225": 107,
         "p362": 108
     },
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "en_GB",
+        "family": "en",
+        "region": "GB"
+    },
+    "dataset": "vctk"
 }

en/en_US/amy/low/en_US-amy-low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "low"
     },
     "espeak": {
         "voice": "en-us"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "en_US",
+        "family": "en",
+        "region": "US"
+    },
+    "dataset": "amy"
 }

en/en_US/amy/medium/en_US-amy-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "en-us"
@@ -479,5 +480,11 @@
     "num_symbols": 256,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "en_US",
+        "family": "en",
+        "region": "US"
+    },
+    "dataset": "amy"
 }

en/en_US/arctic/medium/en_US-arctic-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "en-us"
@@ -498,5 +499,11 @@
         "eey": 16,
         "gka": 17
     },
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "en_US",
+        "family": "en",
+        "region": "US"
+    },
+    "dataset": "arctic"
 }

en/en_US/danny/low/en_US-danny-low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "low"
     },
     "espeak": {
         "voice": "en-us"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "en_US",
+        "family": "en",
+        "region": "US"
+    },
+    "dataset": "danny"
 }

en/en_US/joe/medium/en_US-joe-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "en-us"
@@ -470,5 +471,11 @@
     "num_symbols": 256,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "en_US",
+        "family": "en",
+        "region": "US"
+    },
+    "dataset": "joe"
 }

en/en_US/kathleen/low/en_US-kathleen-low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "low"
     },
     "espeak": {
         "voice": "en-us"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "en_US",
+        "family": "en",
+        "region": "US"
+    },
+    "dataset": "kathleen"
 }

en/en_US/kusal/medium/en_US-kusal-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "en-us"
@@ -479,5 +480,11 @@
     "num_symbols": 256,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "en_US",
+        "family": "en",
+        "region": "US"
+    },
+    "dataset": "kusal"
 }

en/en_US/l2arctic/medium/en_US-l2arctic-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "en-us"
@@ -504,5 +505,11 @@
         "EBVS": 22,
         "SKA": 23
     },
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "en_US",
+        "family": "en",
+        "region": "US"
+    },
+    "dataset": "l2arctic"
 }

en/en_US/lessac/high/en_US-lessac-high.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "high"
     },
     "espeak": {
         "voice": "en-us"
@@ -479,5 +480,11 @@
     "num_symbols": 256,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "en_US",
+        "family": "en",
+        "region": "US"
+    },
+    "dataset": "lessac"
 }

en/en_US/lessac/low/en_US-lessac-low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "low"
     },
     "espeak": {
         "voice": "en-us"
@@ -479,5 +480,11 @@
     "num_symbols": 256,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "en_US",
+        "family": "en",
+        "region": "US"
+    },
+    "dataset": "lessac"
 }

en/en_US/lessac/medium/en_US-lessac-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "en-us"
@@ -479,5 +480,11 @@
     "num_symbols": 256,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "en_US",
+        "family": "en",
+        "region": "US"
+    },
+    "dataset": "lessac"
 }

en/en_US/libritts/high/en_US-libritts-high.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "high"
     },
     "espeak": {
         "voice": "en-us"
@@ -1311,5 +1312,11 @@
         "4681": 902,
         "2085": 903
     },
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "en_US",
+        "family": "en",
+        "region": "US"
+    },
+    "dataset": "libritts"
 }

en/en_US/ryan/high/en_US-ryan-high.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "high"
     },
     "espeak": {
         "voice": "en-us"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "en_US",
+        "family": "en",
+        "region": "US"
+    },
+    "dataset": "ryan"
 }

en/en_US/ryan/low/en_US-ryan-low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "low"
     },
     "espeak": {
         "voice": "en-us"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "en_US",
+        "family": "en",
+        "region": "US"
+    },
+    "dataset": "ryan"
 }

en/en_US/ryan/medium/en_US-ryan-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "en-us"
@@ -479,5 +480,11 @@
     "num_symbols": 256,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "en_US",
+        "family": "en",
+        "region": "US"
+    },
+    "dataset": "ryan"
 }

es/es_ES/carlfm/x_low/es_ES-carlfm-x_low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "x_low"
     },
     "espeak": {
         "voice": "es"
@@ -405,5 +406,12 @@
     },
     "num_symbols": 130,
     "num_speakers": 1,
-    "speaker_id_map": {}
+    "speaker_id_map": {},
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "es_ES",
+        "family": "es",
+        "region": "ES"
+    },
+    "dataset": "carlfm"
 }

es/es_ES/davefx/medium/es_ES-davefx-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "es"
@@ -473,5 +474,11 @@
     "num_symbols": 256,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "es_ES",
+        "family": "es",
+        "region": "ES"
+    },
+    "dataset": "davefx"
 }

es/es_ES/mls_10246/low/es_ES-mls_10246-low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "low"
     },
     "espeak": {
         "voice": "es"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "es_ES",
+        "family": "es",
+        "region": "ES"
+    },
+    "dataset": "mls_10246"
 }

es/es_ES/{mls_9772 → mls_9972}/low/es_ES-mls_9972-low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "low"
     },
     "espeak": {
         "voice": "es"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "es_ES",
+        "family": "es",
+        "region": "ES"
+    },
+    "dataset": "mls_9972"
 }

es/es_ES/sharvard/medium/es_ES-sharvard-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "es"
@@ -482,5 +483,11 @@
         "M": 0,
         "F": 1
     },
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "es_ES",
+        "family": "es",
+        "region": "ES"
+    },
+    "dataset": "sharvard"
 }

es/es_MX/ald/medium/es_MX-ald-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "es-419"
@@ -479,5 +480,11 @@
     "num_symbols": 256,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "es_MX",
+        "family": "es",
+        "region": "MX"
+    },
+    "dataset": "ald"
 }

fi/fi_FI/harri/low/fi_FI-harri-low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "low"
     },
     "espeak": {
         "voice": "fi"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "fi_FI",
+        "family": "fi",
+        "region": "FI"
+    },
+    "dataset": "harri"
 }

fi/fi_FI/harri/medium/fi_FI-harri-medium.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 22050
+        "sample_rate": 22050,
+        "quality": "medium"
     },
     "espeak": {
         "voice": "fi"
@@ -479,5 +480,11 @@
     "num_symbols": 256,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "1.0.0"
+    "piper_version": "1.0.0",
+    "language": {
+        "code": "fi_FI",
+        "family": "fi",
+        "region": "FI"
+    },
+    "dataset": "harri"
 }

fr/fr_FR/gilles/low/fr_FR-gilles-low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "low"
     },
     "espeak": {
         "voice": "fr"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "fr_FR",
+        "family": "fr",
+        "region": "FR"
+    },
+    "dataset": "gilles"
 }

fr/fr_FR/mls_1840/low/fr_FR-mls_1840-low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "low"
     },
     "espeak": {
         "voice": "fr"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "fr_FR",
+        "family": "fr",
+        "region": "FR"
+    },
+    "dataset": "mls_1840"
 }

fr/fr_FR/siwis/low/fr_FR-siwis-low.onnx.json

@@ -1,6 +1,7 @@
 {
     "audio": {
-        "sample_rate": 16000
+        "sample_rate": 16000,
+        "quality": "low"
     },
     "espeak": {
         "voice": "fr"
@@ -406,5 +407,11 @@
     "num_symbols": 130,
     "num_speakers": 1,
     "speaker_id_map": {},
-    "piper_version": "0.2.0"
+    "piper_version": "0.2.0",
+    "language": {
+        "code": "fr_FR",
+        "family": "fr",
+        "region": "FR"
+    },
+    "dataset": "siwis"
 }