Add support for HuggingFace transformers

Author: synesthesiam

CHANGELOG.md

@@ -1,5 +1,9 @@
 # Changelog
 
+## 2.5.0
+
+- Add support for HuggingFace transformers Whisper models (--use-transformers)
+
 ## 2.4.0
 
 - Add "auto" for model and beam size (0) to select values based on CPU

MANIFEST.in

@@ -0,0 +1,73 @@
+[project]
+name = "wyoming-faster-whisper"
+version = "2.5.0"
+description = "Wyoming Server for Faster Whisper"
+readme = "README.md"
+requires-python = ">=3.8.1,<3.13"
+license = {text = "MIT"}
+authors = [
+    {name = "Michael Hansen", email = "[email protected]"}
+]
+keywords = ["rhasspy", "wyoming", "whisper", "stt"]
+classifiers = [
+    "Development Status :: 3 - Alpha",
+    "Intended Audience :: Developers",
+    "Topic :: Text Processing :: Linguistic",
+    "License :: OSI Approved :: MIT License",
+    "Programming Language :: Python :: 3.7",
+    "Programming Language :: Python :: 3.8",
+    "Programming Language :: Python :: 3.9",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+]
+dependencies = [
+    "wyoming>=1.5.3",
+    "faster-whisper>=1.1.0,<2"
+]
+
+[project.urls]
+Homepage = "http://github.com/rhasspy/wyoming-faster-whisper"
+
+[tool.setuptools]
+platforms = ["any"]
+zip-safe  = true
+include-package-data = true
+
+[tool.setuptools.packages.find]
+include = ["wyoming_faster_whisper"]
+exclude = ["tests"]
+
+[project.scripts]
+wyoming-faster-whisper = "wyoming_faster_whisper.__main__:run"
+
+[build-system]
+requires = ["setuptools>=42", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[tool.black]
+line-length = 88
+
+[tool.isort]
+profile = "black"
+
+[tool.pytest.ini_options]
+asyncio_mode = "auto"
+
+[tool.mypy]
+check_untyped_defs = true
+disallow_untyped_defs = true
+
+[project.optional-dependencies]
+dev = [
+    "black==22.12.0",
+    "flake8==6.0.0",
+    "isort==5.11.3",
+    "mypy==0.991",
+    "pylint==2.15.9",
+    "pytest==7.4.4",
+    "pytest-asyncio==0.23.3",
+]
+transformers = [
+    "transformers[torch]==4.52.4",
+]

pyproject.toml

@@ -10,6 +10,9 @@ _VENV_DIR = _PROGRAM_DIR / ".venv"
 
 parser = argparse.ArgumentParser()
 parser.add_argument("--dev", action="store_true", help="Install dev requirements")
+parser.add_argument(
+    "--transformers", action="store_true", help="Install transformers requirements"
+)
 args = parser.parse_args()
 
 # Create virtual environment
@@ -22,11 +25,16 @@ pip = [context.env_exe, "-m", "pip"]
 subprocess.check_call(pip + ["install", "--upgrade", "pip"])
 subprocess.check_call(pip + ["install", "--upgrade", "setuptools", "wheel"])
 
-# Install requirements
-subprocess.check_call(pip + ["install", "-r", str(_PROGRAM_DIR / "requirements.txt")])
-
+extras = []
 if args.dev:
-    # Install dev requirements
-    subprocess.check_call(
-        pip + ["install", "-r", str(_PROGRAM_DIR / "requirements_dev.txt")]
-    )
+    extras.append("dev")
+
+if args.transformers:
+    extras.append("transformers")
+
+extras_str = ""
+if extras:
+    extras_str = "[" + ",".join(extras) + "]"
+
+# Install requirements
+subprocess.check_call(pip + ["install", "-e", f"{_PROGRAM_DIR}{extras_str}"])

requirements.txt

@@ -5,6 +5,7 @@
 import platform
 import re
 from functools import partial
+from typing import Any
 
 import faster_whisper
 from wyoming.info import AsrModel, AsrProgram, Attribution, Info
@@ -59,6 +60,16 @@ async def main() -> None:
         "--initial-prompt",
         help="Optional text to provide as a prompt for the first window",
     )
+    parser.add_argument(
+        "--use-transformers",
+        action="store_true",
+        help="Use HuggingFace transformers library (requires transformers extras)",
+    )
+    parser.add_argument(
+        "--local-files-only",
+        action="store_true",
+        help="Don't check HuggingFace hub for updates every time",
+    )
     #
     parser.add_argument("--debug", action="store_true", help="Log DEBUG messages")
     parser.add_argument(
@@ -135,26 +146,61 @@ async def main() -> None:
 
     # Load model
     _LOGGER.debug("Loading %s", args.model)
-    whisper_model = faster_whisper.WhisperModel(
-        args.model,
-        download_root=args.download_dir,
-        device=args.device,
-        compute_type=args.compute_type,
-    )
+    whisper_model: Any = None
+
+    if args.use_transformers:
+        # Use HuggingFace transformers
+        from .transformers_whisper import TransformersWhisperModel
+
+        whisper_model = TransformersWhisperModel(
+            args.model, args.download_dir, args.local_files_only
+        )
+    else:
+        # Use faster-whisper
+        whisper_model = faster_whisper.WhisperModel(
+            args.model,
+            download_root=args.download_dir,
+            device=args.device,
+            compute_type=args.compute_type,
+        )
 
     server = AsyncServer.from_uri(args.uri)
     _LOGGER.info("Ready")
     model_lock = asyncio.Lock()
-    await server.run(
-        partial(
-            FasterWhisperEventHandler,
-            wyoming_info,
-            args,
-            whisper_model,
-            model_lock,
-            initial_prompt=args.initial_prompt,
+
+    if args.use_transformers:
+        # Use HuggingFace transformers
+        from .transformers_whisper import (
+            TransformersWhisperEventHandler,
+            TransformersWhisperModel,
+        )
+
+        assert isinstance(whisper_model, TransformersWhisperModel)
+
+        # TODO: initial prompt
+        await server.run(
+            partial(
+                TransformersWhisperEventHandler,
+                wyoming_info,
+                args.language,
+                args.beam_size,
+                whisper_model,
+                model_lock,
+            )
+        )
+    else:
+        # Use faster-whisper
+        assert isinstance(whisper_model, faster_whisper.WhisperModel)
+        await server.run(
+            partial(
+                FasterWhisperEventHandler,
+                wyoming_info,
+                args,
+                whisper_model,
+                model_lock,
+                initial_prompt=args.initial_prompt,
+            )
         )
-    )
 
 
 # -----------------------------------------------------------------------------