Add multiprocess mechanism for Common Voice

lhotse/__init__.py

@@ -4,6 +4,7 @@
     RecordingSet,
     set_audio_duration_mismatch_tolerance,
     set_ffmpeg_torchaudio_info_enabled,
+    get_ffmpeg_torchaudio_info_enabled,
 )
 from .caching import is_caching_enabled, set_caching_enabled
 from .cut import CutSet, MonoCut, MultiCut, create_cut_set_eager, create_cut_set_lazy

lhotse/audio.py

@@ -143,6 +143,19 @@ def set_ffmpeg_torchaudio_info_enabled(enabled: bool) -> None:
     FFMPEG_TORCHAUDIO_INFO_ENABLED = enabled
 
 
+def get_ffmpeg_torchaudio_info_enabled() -> bool:
+    """
+    Return FFMPEG_TORCHAUDIO_INFO_ENABLED, which is Lhotse's global setting for whether to
+    use ffmpeg-torchaudio to compute the duration of audio files.
+
+    Example::
+
+        >>> import lhotse
+        >>> lhotse.get_ffmpeg_torchaudio_info_enabled()
+    """
+    return FFMPEG_TORCHAUDIO_INFO_ENABLED
+
+
 # TODO: document the dataclasses like this:
 # https://stackoverflow.com/a/3051356/5285891
 

lhotse/bin/modes/recipes/commonvoice.py

@@ -64,8 +64,8 @@ def commonvoice(
 @click.option(
     "-v",
     "--release",
-    default="cv-corpus-8.0-2022-01-19",
-    help="the name of the CommonVoice release (e.g., 'cv-corpus-8.0-2022-01-19').It is used as part of the download URL.",
+    default="cv-corpus-13.0-2023-03-09",
+    help="the name of the CommonVoice release (e.g., 'cv-corpus-13.0-2023-03-09').It is used as part of the download URL.",
 )
 def commonvoice(
     target_dir: Pathlike,