add private module to create model object for speech api

opendbm/api_lib/speech/_transcribe.py

@@ -0,0 +1,16 @@
+import tempfile
+
+from opendbm.api_lib.model import OPENDBM_DATA, AudioModel
+from opendbm.dbm_lib.dbm_features.raw_features.nlp.transcribe import run_transcribe
+
+
+class Transcribe(AudioModel):
+    def __init__(self):
+        super().__init__()
+        self._params = ["nlp_transcribe", "nlp_totalTime"]
+
+    @AudioModel.prep_func
+    def _fit_transform(self, path):
+        return run_transcribe(
+            path, f"{tempfile.gettempdir()}/", self.r_config, OPENDBM_DATA, save=False
+        )