wip: add pyannote.audio CLI

pyannote/audio/__main__.py

@@ -0,0 +1,72 @@
+# MIT License
+#
+# Copyright (c) 2022- CNRS
+#
+# Permission is hereby granted, free of charge, to any person obtaining a copy
+# of this software and associated documentation files (the "Software"), to deal
+# in the Software without restriction, including without limitation the rights
+# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+# copies of the Software, and to permit persons to whom the Software is
+# furnished to do so, subject to the following conditions:
+#
+# The above copyright notice and this permission notice shall be included in all
+# copies or substantial portions of the Software.
+#
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+# SOFTWARE.
+
+
+from pathlib import Path
+
+import typer
+from pyannote.core import Annotation
+
+from pyannote.audio import Audio, Pipeline
+from pyannote.audio.pipelines.utils.hook import Hooks, ProgressHook, TimingHook
+
+app = typer.Typer()
+
+
+@app.command("apply")
+def apply(
+    pipeline: str,
+    audio_in: str,
+    rttm_out: str,
+):
+    pretrained_pipeline = Pipeline.from_pretrained(pipeline)
+
+    io = Audio()
+
+    with open(rttm_out, "w") as rttm:
+        audio_duration = io.get_duration(audio_in)
+
+        file = {
+            "audio": audio_in,
+            "uri": Path(audio_in).stem,
+        }
+
+        rttm.write(f"# pipeline.path: {pipeline}\n")
+        rttm.write(f"# audio.path: {audio_in}\n")
+        rttm.write(f"# audio.duration: {audio_duration:.1f}s\n")
+
+        with Hooks(ProgressHook(), TimingHook()) as hook:
+            output: Annotation = pretrained_pipeline(file, hook=hook)
+
+        time_total = file["timing_hook"]["total"]
+
+        rttm.write(f"# processing.duration: {time_total:.1f}s\n")
+        rttm.write(f"# processing.rtf: {time_total / audio_duration:.0%}\n")
+
+        output.write_rttm(rttm)
+        rttm.flush()
+
+
+if __name__ == "__main__":
+    app()
+
+# python -m pyannote.audio apply pyannote/speaker-diarization-3.0 file.wav file.wav.rttm