Create OpenAIWhisperParser for generating Documents from audio files (#5580)

# OpenAIWhisperParser

This PR creates a new parser, `OpenAIWhisperParser`, that uses the
[OpenAI Whisper
model](https://platform.openai.com/docs/guides/speech-to-text/quickstart)
to perform transcription of audio files to text (`Documents`). Please
see the notebook for usage.

docs/modules/indexes/document_loaders.rst

@@ -30,6 +30,7 @@ For detailed instructions on how to get set up with Unstructured, see installati
    :maxdepth: 1
    :glob:
 
+   ./document_loaders/examples/audio.ipynb
    ./document_loaders/examples/conll-u.ipynb
    ./document_loaders/examples/copypaste.ipynb
    ./document_loaders/examples/csv.ipynb

langchain/document_loaders/parsers/__init__.py

@@ -1,3 +1,4 @@
+from langchain.document_loaders.parsers.audio import OpenAIWhisperParser
 from langchain.document_loaders.parsers.html import BS4HTMLParser
 from langchain.document_loaders.parsers.pdf import (
     PDFMinerParser,
@@ -9,6 +10,7 @@ from langchain.document_loaders.parsers.pdf import (
 
 __all__ = [
     "BS4HTMLParser",
+    "OpenAIWhisperParser",
     "PDFMinerParser",
     "PDFPlumberParser",
     "PyMuPDFParser",

langchain/document_loaders/parsers/audio.py

@@ -0,0 +1,21 @@
+from typing import Iterator
+
+from langchain.document_loaders.base import BaseBlobParser
+from langchain.document_loaders.blob_loaders import Blob
+from langchain.schema import Document
+
+
+class OpenAIWhisperParser(BaseBlobParser):
+    """Transcribe and parse audio files.
+    Audio transcription is with OpenAI Whisper model."""
+
+    def lazy_parse(self, blob: Blob) -> Iterator[Document]:
+        """Lazily parse the blob."""
+
+        import openai
+
+        with blob.as_bytes_io() as f:
+            transcript = openai.Audio.transcribe("whisper-1", f)
+            yield Document(
+                page_content=transcript.text, metadata={"source": blob.source}
+            )

tests/unit_tests/document_loaders/parsers/test_public_api.py

@@ -5,6 +5,7 @@ def test_parsers_public_api_correct() -> None:
     """Test public API of parsers for breaking changes."""
     assert set(__all__) == {
         "BS4HTMLParser",
+        "OpenAIWhisperParser",
         "PyPDFParser",
         "PDFMinerParser",
         "PyMuPDFParser",