chore: update schemas

EveryVoiceTTS · Apr 16, 2024 · bac8d0a · bac8d0a
1 parent f5a8186
commit bac8d0a
Show file tree

Hide file tree

Showing 6 changed files with 790 additions and 144 deletions.
diff --git a/everyvoice/.schema/everyvoice-aligner-schema-0.1.json b/everyvoice/.schema/everyvoice-aligner-schema-0.1.json
@@ -250,6 +250,14 @@
       "$schema": "http://json-schema.org/draft-07/schema#",
       "additionalProperties": false,
       "properties": {
+        "target_text_representation_level": {
+          "allOf": [
+            {
+              "$ref": "#/$defs/TargetTrainingTextRepresentationLevel"
+            }
+          ],
+          "default": "characters"
+        },
         "lstm_dim": {
           "default": 512,
           "description": "The number of dimensions in the LSTM layers.",
@@ -477,7 +485,7 @@
         "save_dir": {
           "default": "logs_and_checkpoints",
           "description": "The directory to save your checkpoints and logs to.",
-          "format": "directory-path",
+          "format": "path",
           "title": "Save Dir",
           "type": "string"
         },
@@ -523,7 +531,7 @@
         "save_dir": {
           "default": "preprocessed/YourDataSet",
           "description": "The directory to save preprocessed files to.",
-          "format": "directory-path",
+          "format": "path",
           "title": "Save Dir",
           "type": "string"
         },
@@ -561,53 +569,124 @@
       "title": "PreprocessingConfig",
       "type": "object"
     },
+    "Punctuation": {
+      "properties": {
+        "exclamations": {
+          "default": [
+            "!",
+            "\u00a1"
+          ],
+          "description": "Exclamation punctuation symbols used in your datasets. Replaces these symbols with <EXCL> internally.",
+          "items": {
+            "type": "string"
+          },
+          "title": "Exclamations",
+          "type": "array"
+        },
+        "question_symbols": {
+          "default": [
+            "?",
+            "\u00bf"
+          ],
+          "description": "Question/interrogative punctuation symbols used in your datasets. Replaces these symbols with <QINT> internally.",
+          "items": {
+            "type": "string"
+          },
+          "title": "Question Symbols",
+          "type": "array"
+        },
+        "quotemarks": {
+          "default": [
+            "\"",
+            "'",
+            "\u201c",
+            "\u201d",
+            "\u00ab",
+            "\u00bb"
+          ],
+          "description": "Quotemark punctuation symbols used in your datasets. Replaces these symbols with <QUOTE> internally.",
+          "items": {
+            "type": "string"
+          },
+          "title": "Quotemarks",
+          "type": "array"
+        },
+        "big_breaks": {
+          "default": [
+            ".",
+            ":",
+            ";"
+          ],
+          "description": "Punctuation symbols indicating a 'big break' used in your datasets. Replaces these symbols with <BB> internally.",
+          "items": {
+            "type": "string"
+          },
+          "title": "Big Breaks",
+          "type": "array"
+        },
+        "small_breaks": {
+          "default": [
+            ",",
+            "-",
+            "\u2014"
+          ],
+          "description": "Punctuation symbols indicating a 'small break' used in your datasets. Replaces these symbols with <SB> internally.",
+          "items": {
+            "type": "string"
+          },
+          "title": "Small Breaks",
+          "type": "array"
+        },
+        "ellipsis": {
+          "default": [
+            "\u2026"
+          ],
+          "description": "Punctuation symbols indicating an ellipsis used in your datasets. Replaces these symbols with <EPS> internally.",
+          "items": {
+            "type": "string"
+          },
+          "title": "Ellipsis",
+          "type": "array"
+        }
+      },
+      "title": "Punctuation",
+      "type": "object"
+    },
     "Symbols": {
       "additionalProperties": true,
       "properties": {
         "silence": {
-          "anyOf": [
-            {
-              "type": "string"
-            },
-            {
-              "items": {
-                "type": "string"
-              },
-              "type": "array"
-            }
-          ],
           "default": [
             "<SIL>"
           ],
           "description": "The symbol(s) used to indicate silence.",
-          "title": "Silence"
-        },
-        "pad": {
-          "default": "_",
-          "description": "The symbol used to indicate padding. Batches are length-normalized by adding this padding character so that each utterance in the batch is the same length.",
-          "title": "Pad",
-          "type": "string"
+          "items": {
+            "type": "string"
+          },
+          "title": "Silence",
+          "type": "array"
         },
         "punctuation": {
-          "anyOf": [
-            {
-              "type": "string"
-            },
+          "allOf": [
             {
-              "items": {
-                "type": "string"
-              },
-              "type": "array"
+              "$ref": "#/$defs/Punctuation"
             }
           ],
-          "default": "-';:,.!?\u00a1\u00bf\u2014\u2026\"\u00ab\u00bb\u201c\u201d ",
-          "description": "A list of punctuation symbols.",
-          "title": "Punctuation"
+          "description": "EveryVoice will combine punctuation and normalize it into a set of five permissible types of punctuation to help tractable training."
         }
       },
       "title": "Symbols",
       "type": "object"
     },
+    "TargetTrainingTextRepresentationLevel": {
+      "enum": [
+        "characters",
+        "phones",
+        "phonological_features"
+      ],
+      "title": "TargetTrainingTextRepresentationLevel",
+      "type": "string"
+    },
     "TextConfig": {
       "$schema": "http://json-schema.org/draft-07/schema#",
       "additionalProperties": false,

diff --git a/everyvoice/.schema/everyvoice-shared-data-schema-0.1.json b/everyvoice/.schema/everyvoice-shared-data-schema-0.1.json
@@ -187,7 +187,7 @@
     "save_dir": {
       "default": "preprocessed/YourDataSet",
       "description": "The directory to save preprocessed files to.",
-      "format": "directory-path",
+      "format": "path",
       "title": "Save Dir",
       "type": "string"
     },

diff --git a/everyvoice/.schema/everyvoice-shared-text-schema-0.1.json b/everyvoice/.schema/everyvoice-shared-text-schema-0.1.json
@@ -1,47 +1,109 @@
 {
   "$defs": {
+    "Punctuation": {
+      "properties": {
+        "exclamations": {
+          "default": [
+            "!",
+            "\u00a1"
+          ],
+          "description": "Exclamation punctuation symbols used in your datasets. Replaces these symbols with <EXCL> internally.",
+          "items": {
+            "type": "string"
+          },
+          "title": "Exclamations",
+          "type": "array"
+        },
+        "question_symbols": {
+          "default": [
+            "?",
+            "\u00bf"
+          ],
+          "description": "Question/interrogative punctuation symbols used in your datasets. Replaces these symbols with <QINT> internally.",
+          "items": {
+            "type": "string"
+          },
+          "title": "Question Symbols",
+          "type": "array"
+        },
+        "quotemarks": {
+          "default": [
+            "\"",
+            "'",
+            "\u201c",
+            "\u201d",
+            "\u00ab",
+            "\u00bb"
+          ],
+          "description": "Quotemark punctuation symbols used in your datasets. Replaces these symbols with <QUOTE> internally.",
+          "items": {
+            "type": "string"
+          },
+          "title": "Quotemarks",
+          "type": "array"
+        },
+        "big_breaks": {
+          "default": [
+            ".",
+            ":",
+            ";"
+          ],
+          "description": "Punctuation symbols indicating a 'big break' used in your datasets. Replaces these symbols with <BB> internally.",
+          "items": {
+            "type": "string"
+          },
+          "title": "Big Breaks",
+          "type": "array"
+        },
+        "small_breaks": {
+          "default": [
+            ",",
+            "-",
+            "\u2014"
+          ],
+          "description": "Punctuation symbols indicating a 'small break' used in your datasets. Replaces these symbols with <SB> internally.",
+          "items": {
+            "type": "string"
+          },
+          "title": "Small Breaks",
+          "type": "array"
+        },
+        "ellipsis": {
+          "default": [
+            "\u2026"
+          ],
+          "description": "Punctuation symbols indicating an ellipsis used in your datasets. Replaces these symbols with <EPS> internally.",
+          "items": {
+            "type": "string"
+          },
+          "title": "Ellipsis",
+          "type": "array"
+        }
+      },
+      "title": "Punctuation",
+      "type": "object"
+    },
     "Symbols": {
       "additionalProperties": true,
       "properties": {
         "silence": {
-          "anyOf": [
-            {
-              "type": "string"
-            },
-            {
-              "items": {
-                "type": "string"
-              },
-              "type": "array"
-            }
-          ],
           "default": [
             "<SIL>"
           ],
           "description": "The symbol(s) used to indicate silence.",
-          "title": "Silence"
-        },
-        "pad": {
-          "default": "_",
-          "description": "The symbol used to indicate padding. Batches are length-normalized by adding this padding character so that each utterance in the batch is the same length.",
-          "title": "Pad",
-          "type": "string"
+          "items": {
+            "type": "string"
+          },
+          "title": "Silence",
+          "type": "array"
         },
         "punctuation": {
-          "anyOf": [
-            {
-              "type": "string"
-            },
+          "allOf": [
             {
-              "items": {
-                "type": "string"
-              },
-              "type": "array"
+              "$ref": "#/$defs/Punctuation"
             }
           ],
-          "default": "-';:,.!?\u00a1\u00bf\u2014\u2026\"\u00ab\u00bb\u201c\u201d ",
-          "description": "A list of punctuation symbols.",
-          "title": "Punctuation"
+          "description": "EveryVoice will combine punctuation and normalize it into a set of five permissible types of punctuation to help tractable training."
         }
       },
       "title": "Symbols",

diff --git a/everyvoice/.schema/everyvoice-spec-to-wav-schema-0.1.json b/everyvoice/.schema/everyvoice-spec-to-wav-schema-0.1.json
@@ -600,7 +600,7 @@
         "save_dir": {
           "default": "logs_and_checkpoints",
           "description": "The directory to save your checkpoints and logs to.",
-          "format": "directory-path",
+          "format": "path",
           "title": "Save Dir",
           "type": "string"
         },
@@ -646,7 +646,7 @@
         "save_dir": {
           "default": "preprocessed/YourDataSet",
           "description": "The directory to save preprocessed files to.",
-          "format": "directory-path",
+          "format": "path",
           "title": "Save Dir",
           "type": "string"
         },