bootphon · thewh1teagle · Jan 4, 2025
diff --git a/phonemizer/backend/espeak/api.py b/phonemizer/backend/espeak/api.py
@@ -40,10 +40,13 @@ class EspeakAPI:
 
     """
 
-    def __init__(self, library: Union[str, Path]):
+    def __init__(self, library: Union[str, Path], data_path: Union[str, Path, None]):
         # set to None to avoid an AttributeError in _delete if the __init__
         # method raises, will be properly initialized below
         self._library = None
+
+        if data_path is not None:
+            data_path = str(data_path).encode('utf-8')
 
         # Because the library is not designed to be wrapped nor to be used in
         # multithreaded/multiprocess contexts (massive use of global variables)
@@ -83,7 +86,7 @@ def __init__(self, library: Union[str, Path]):
         # AUDIO_OUTPUT_SYNCHRONOUS in the espeak API
         self._library = ctypes.cdll.LoadLibrary(str(espeak_copy))
         try:
-            if self._library.espeak_Initialize(0x02, 0, None, 0) <= 0:
+            if self._library.espeak_Initialize(0x02, 0, data_path, 0) <= 0:
                 raise RuntimeError(  # pragma: nocover
                     'failed to initialize espeak shared library')
         except AttributeError:  # pragma: nocover

diff --git a/phonemizer/backend/espeak/wrapper.py b/phonemizer/backend/espeak/wrapper.py
@@ -48,6 +48,7 @@ class EspeakWrapper:
     # on the system. The user can choose an alternative espeak library with
     # the method EspeakWrapper.set_library().
     _ESPEAK_LIBRARY = None
+    _ESPEAK_DATA_PATH = None
 
     def __init__(self):
         # the following attributes are accessed through properties and are
@@ -57,7 +58,7 @@ def __init__(self):
         self._voice = None
 
         # load the espeak API
-        self._espeak = EspeakAPI(self.library())
+        self._espeak = EspeakAPI(self.library(), self.data_path)
 
         # lazy loading of attributes only required for the synthetize method
         self._libc_ = None
@@ -113,6 +114,21 @@ def set_library(cls, library: str):
 
         """
         cls._ESPEAK_LIBRARY = library
+
+    @classmethod
+    def set_data_path(cls, data_path: str):
+        """Sets the path for the data to be used by the espeak backend.
+
+        If this is not set, the backend uses the default data path from the system installation.
+
+        Parameters
+        ----------
+        data_path : str
+            The path to the data to be used by the espeak backend. Set `data_path` to None
+            to restore the default.
+
+        """
+        cls._ESPEAK_DATA_PATH = data_path
 
     @classmethod
     def library(cls):
@@ -179,8 +195,37 @@ def library_path(self):
 
     @property
     def data_path(self):
-        """The espeak data directory as a pathlib.Path instance"""
-        if self._data_path is None:
+        """Returns the espeak library used as backend
+
+        The following precedence rule applies for library lookup:
+
+        1. As specified by BaseEspeakBackend.set_library()
+        2. Or as specified by the environment variable
+           PHONEMIZER_ESPEAK_LIBRARY
+        3. Or the default espeak library found on the system
+
+        Raises
+        ------
+        RuntimeError if the espeak library cannot be found or if the
+          environment variable PHONEMIZER_ESPEAK_LIBRARY is set to a
+          non-readable file
+
+        """
+        if self._ESPEAK_DATA_PATH:
+            data_path = pathlib.Path(self._ESPEAK_DATA_PATH)
+            if not (data_path.is_dir() and os.access(self._ESPEAK_DATA_PATH, os.R_OK)):
+                raise RuntimeError(f'{self._ESPEAK_DATA_PATH} is not a readable directory')
+            self._data_path = data_path.resolve()
+        elif 'PHONEMIZER_ESPEAK_DATA_PATH' in os.environ:
+            data_path = pathlib.Path(os.environ['PHONEMIZER_ESPEAK_DATA_PATH'])
+            if not (data_path.is_dir() and os.access(data_path, os.R_OK)):
+                raise RuntimeError(  # pragma: nocover
+                    f'PHONEMIZER_ESPEAK_DATA_PATH={data_path} '
+                    f'is not a readable directory')
+            self._data_path = data_path.resolve()
+
+        # Fetch path dynamically after initialize
+        if self._data_path is None and hasattr(self, '_espeak'):
             self._fetch_version_and_path()
         return self._data_path