Add recipe for the Santa Barbara Corpus of Spoken American English (SBCSAE)

docs/corpus.rst

@@ -173,6 +173,8 @@ a CLI tool that create the manifests given a corpus directory.
     - :func:`lhotse.recipes.prepare_reazonspeech`
   * - RIRs and Noises Corpus (OpenSLR 28)
     - :func:`lhotse.recipes.prepare_rir_noise`
+  * - SBCSAE
+    - :func:`lhotse.recipes.prepare_sbcsae`
   * - Spatial-LibriSpeech
     - :func:`lhotse.recipes.prepare_spatial_librispeech`
   * - Speech Commands

lhotse/bin/modes/recipes/__init__.py

@@ -67,6 +67,7 @@
 from .primewords import *
 from .reazonspeech import *
 from .rir_noise import *
+from .sbcsae import *
 from .slu import *
 from .spatial_librispeech import *
 from .speechcommands import *

lhotse/bin/modes/recipes/sbcsae.py

@@ -0,0 +1,58 @@
+from typing import Optional, Sequence
+
+import click
+
+from lhotse.bin.modes import download, prepare
+from lhotse.recipes.sbcsae import download_sbcsae, prepare_sbcsae
+from lhotse.utils import Pathlike
+
+__all__ = ["sbcsae"]
+
+
+@prepare.command(context_settings=dict(show_default=True))
+@click.argument("corpus_dir", type=click.Path(exists=True, dir_okay=True))
+@click.argument("output_dir", type=click.Path())
+@click.option(
+    "--geolocation",
+    type=bool,
+    is_flag=True,
+    default=False,
+    help="Include geographic coordinates of speakers' hometowns in the manifests.",
+)
+@click.option(
+    "--omit-realignments",
+    type=bool,
+    is_flag=True,
+    default=False,
+    help="Only output the original corpus segmentation without boundary improvements.",
+)
+def sbcsae(
+    corpus_dir: Pathlike,
+    output_dir: Pathlike,
+    geolocation: bool,
+    omit_realignments: bool,
+):
+    """SBCSAE data preparation."""
+    prepare_sbcsae(
+        corpus_dir,
+        output_dir=output_dir,
+        geolocation=geolocation,
+        omit_realignments=omit_realignments,
+    )
+
+
+@download.command(context_settings=dict(show_default=True))
+@click.argument("target_dir", type=click.Path())
+@click.option(
+    "--force-download",
+    type=bool,
+    is_flag=True,
+    default=False,
+    help="Force download.",
+)
+def sbcsae(
+    target_dir: Pathlike,
+    force_download: bool,
+):
+    """SBCSAE download."""
+    download_sbcsae(target_dir, force_download=force_download)

lhotse/recipes/__init__.py

@@ -68,6 +68,7 @@
 from .peoples_speech import prepare_peoples_speech
 from .reazonspeech import download_reazonspeech, prepare_reazonspeech
 from .rir_noise import download_rir_noise, prepare_rir_noise
+from .sbcsae import download_sbcsae, prepare_sbcsae
 from .slu import prepare_slu
 from .spatial_librispeech import (
     download_spatial_librispeech,