funasr-haystack 0.0.1__tar.gz

funasr_haystack-0.0.1/.gitignore

@@ -0,0 +1,146 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+pip-wheel-metadata/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+volumes/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+.python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# IDEs
+.vscode
+
+# Docs generation artifacts
+_readme_*.md
+.idea
+
+# macOS
+.DS_Store
+
+# http cache (requests-cache)
+**/http_cache.sqlite
+
+# ruff
+.ruff_cache

funasr_haystack-0.0.1/LICENSE.txt

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2023-present deepset GmbH
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

funasr_haystack-0.0.1/PKG-INFO

@@ -0,0 +1,39 @@
+Metadata-Version: 2.4
+Name: funasr-haystack
+Version: 0.0.1
+Summary: FunASR speech-to-text transcriber for Haystack
+Project-URL: Documentation, https://github.com/deepset-ai/haystack-core-integrations/tree/main/integrations/funasr#readme
+Project-URL: Issues, https://github.com/deepset-ai/haystack-core-integrations/issues
+Project-URL: Source, https://github.com/deepset-ai/haystack-core-integrations/tree/main/integrations/funasr
+Author-email: deepset GmbH <info@deepset.ai>
+License-Expression: Apache-2.0
+License-File: LICENSE.txt
+Keywords: asr,audio,funasr,haystack,llm,rag,speech-to-text,transcription
+Classifier: Development Status :: 4 - Beta
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Programming Language :: Python
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
+Classifier: Programming Language :: Python :: Implementation :: CPython
+Requires-Python: >=3.10
+Requires-Dist: funasr>=1.0.0
+Requires-Dist: haystack-ai>=2.9.0
+Requires-Dist: torch
+Requires-Dist: torchaudio
+Description-Content-Type: text/markdown
+
+# funasr-haystack
+
+[![PyPI - Version](https://img.shields.io/pypi/v/funasr-haystack.svg)](https://pypi.org/project/funasr-haystack)
+[![PyPI - Python Version](https://img.shields.io/pypi/pyversions/funasr-haystack.svg)](https://pypi.org/project/funasr-haystack)
+
+- [Changelog](https://github.com/deepset-ai/haystack-core-integrations/blob/main/integrations/funasr/CHANGELOG.md)
+
+---
+
+## Contributing
+
+Refer to the general [Contribution Guidelines](https://github.com/deepset-ai/haystack-core-integrations/blob/main/CONTRIBUTING.md).

funasr_haystack-0.0.1/README.md

@@ -0,0 +1,12 @@
+# funasr-haystack
+
+[![PyPI - Version](https://img.shields.io/pypi/v/funasr-haystack.svg)](https://pypi.org/project/funasr-haystack)
+[![PyPI - Python Version](https://img.shields.io/pypi/pyversions/funasr-haystack.svg)](https://pypi.org/project/funasr-haystack)
+
+- [Changelog](https://github.com/deepset-ai/haystack-core-integrations/blob/main/integrations/funasr/CHANGELOG.md)
+
+---
+
+## Contributing
+
+Refer to the general [Contribution Guidelines](https://github.com/deepset-ai/haystack-core-integrations/blob/main/CONTRIBUTING.md).

funasr_haystack-0.0.1/pydoc/config_docusaurus.yml

@@ -0,0 +1,13 @@
+loaders:
+  - modules:
+      - haystack_integrations.components.audio.funasr.transcriber
+    search_path: [../src]
+processors:
+  - type: filter
+    documented_only: true
+    skip_empty_modules: true
+renderer:
+  description: FunASR speech-to-text integration for Haystack
+  id: integrations-funasr
+  filename: funasr.md
+  title: FunASR

funasr_haystack-0.0.1/pyproject.toml

@@ -0,0 +1,170 @@
+[build-system]
+requires = ["hatchling", "hatch-vcs"]
+build-backend = "hatchling.build"
+
+[project]
+name = "funasr-haystack"
+dynamic = ["version"]
+description = "FunASR speech-to-text transcriber for Haystack"
+readme = "README.md"
+requires-python = ">=3.10"
+license = "Apache-2.0"
+keywords = ["funasr", "speech-to-text", "asr", "audio", "transcription", "haystack", "llm", "rag"]
+authors = [{ name = "deepset GmbH", email = "info@deepset.ai" }]
+classifiers = [
+  "License :: OSI Approved :: Apache Software License",
+  "Development Status :: 4 - Beta",
+  "Programming Language :: Python",
+  "Programming Language :: Python :: 3.10",
+  "Programming Language :: Python :: 3.11",
+  "Programming Language :: Python :: 3.12",
+  "Programming Language :: Python :: 3.13",
+  "Programming Language :: Python :: 3.14",
+  "Programming Language :: Python :: Implementation :: CPython",
+]
+dependencies = [
+    "haystack-ai>=2.9.0",
+    "funasr>=1.0.0",
+    "torch",
+    "torchaudio",
+]
+
+[project.urls]
+Documentation = "https://github.com/deepset-ai/haystack-core-integrations/tree/main/integrations/funasr#readme"
+Issues = "https://github.com/deepset-ai/haystack-core-integrations/issues"
+Source = "https://github.com/deepset-ai/haystack-core-integrations/tree/main/integrations/funasr"
+
+[tool.hatch.build.targets.wheel]
+packages = ["src/haystack_integrations"]
+
+[tool.hatch.version]
+source = "vcs"
+tag-pattern = 'integrations\/funasr-v(?P<version>.*)'
+
+[tool.hatch.version.raw-options]
+root = "../.."
+git_describe_command = 'git describe --tags --match="integrations/funasr-v[0-9]*"'
+
+[tool.hatch.envs.default]
+installer = "uv"
+dependencies = ["haystack-pydoc-tools", "ruff"]
+
+[tool.hatch.envs.default.scripts]
+docs = ["haystack-pydoc pydoc/config_docusaurus.yml"]
+fmt = "ruff check --fix {args}; ruff format {args}"
+fmt-check = "ruff check {args} && ruff format --check {args}"
+
+[tool.hatch.envs.test]
+dependencies = [
+    "pytest",
+    "pytest-asyncio",
+    "pytest-cov",
+    "pytest-rerunfailures",
+    "mypy",
+    "pip",
+    "funasr>=1.0.0",
+]
+
+[tool.hatch.envs.test.scripts]
+unit = 'pytest -m "not integration" {args:tests}'
+integration = 'pytest -m "integration" {args:tests}'
+all = 'pytest {args:tests}'
+unit-cov-retry = 'pytest --cov=haystack_integrations --reruns 3 --reruns-delay 30 -x -m "not integration" {args:tests}'
+integration-cov-append-retry = 'pytest --cov=haystack_integrations --cov-append --reruns 3 --reruns-delay 30 -x -m "integration" {args:tests}'
+types = "mypy -p haystack_integrations.components.audio.funasr {args}"
+
+[tool.mypy]
+install_types = true
+non_interactive = true
+check_untyped_defs = true
+disallow_incomplete_defs = true
+
+[[tool.mypy.overrides]]
+module = ["funasr", "funasr.*"]
+ignore_missing_imports = true
+
+[tool.ruff]
+line-length = 120
+
+[tool.ruff.lint]
+select = [
+    "A",
+    "ANN",
+    "ARG",
+    "B",
+    "C",
+    "D102",   # Missing docstring in public method
+    "D103",   # Missing docstring in public function
+    "D205",   # 1 blank line required between summary line and description
+    "D209",   # Closing triple quotes go to new line
+    "D213",   # summary lines must be positioned on the second physical line of the docstring
+    "D417",   # Missing argument descriptions in the docstring
+    "D419",   # Docstring is empty
+    "DTZ",
+    "E",
+    "EM",
+    "F",
+    "I",
+    "ICN",
+    "ISC",
+    "N",
+    "PLC",
+    "PLE",
+    "PLR",
+    "PLW",
+    "Q",
+    "RUF",
+    "S",
+    "T",
+    "TID",
+    "UP",
+    "W",
+    "YTT",
+]
+ignore = [
+    # Allow non-abstract empty methods in abstract base classes
+    "B027",
+    # Allow function calls in argument defaults (common Haystack pattern for Secret.from_env_var)
+    "B008",
+    # Ignore checks for possible passwords
+    "S105",
+    "S106",
+    "S107",
+    # Ignore complexity
+    "C901",
+    "PLR0911",
+    "PLR0912",
+    "PLR0913",
+    "PLR0915",
+    # Allow `Any` type - used legitimately for dynamic types and SDK boundaries
+    "ANN401",
+]
+
+[tool.ruff.lint.isort]
+known-first-party = ["haystack_integrations"]
+
+[tool.ruff.lint.flake8-tidy-imports]
+ban-relative-imports = "parents"
+
+[tool.ruff.lint.per-file-ignores]
+# Tests can use magic values, assertions, relative imports, and don't need type annotations
+"tests/**/*" = ["PLR2004", "S101", "TID252", "D", "ANN"]
+
+[tool.coverage.run]
+source = ["haystack_integrations"]
+branch = true
+parallel = false
+relative_files = true
+
+[tool.coverage.report]
+omit = ["*/tests/*", "*/__init__.py"]
+show_missing = true
+exclude_lines = ["no cov", "if __name__ == .__main__.:", "if TYPE_CHECKING:"]
+
+[tool.pytest.ini_options]
+addopts = "--strict-markers"
+markers = [
+  "integration: integration tests",
+]
+log_cli = true
+asyncio_default_fixture_loop_scope = "function"

funasr_haystack-0.0.1/src/haystack_integrations/components/audio/funasr/__init__.py

@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: 2022-present deepset GmbH <info@deepset.ai>
+#
+# SPDX-License-Identifier: Apache-2.0
+
+from haystack_integrations.components.audio.funasr.transcriber import FunASRTranscriber
+
+__all__ = ["FunASRTranscriber"]

funasr_haystack-0.0.1/src/haystack_integrations/components/audio/funasr/transcriber.py

@@ -0,0 +1,270 @@
+# SPDX-FileCopyrightText: 2023-present deepset GmbH <info@deepset.ai>
+#
+# SPDX-License-Identifier: Apache-2.0
+
+import contextlib
+import os
+import tempfile
+from pathlib import Path
+from typing import Any
+
+from haystack import Document, component, default_from_dict, default_to_dict, logging
+from haystack.components.converters.utils import normalize_metadata
+from haystack.dataclasses import ByteStream
+from haystack.utils import ComponentDevice
+
+from funasr import AutoModel
+
+logger = logging.getLogger(__name__)
+
+_MIME_TO_SUFFIX: dict[str, str] = {
+    "audio/wav": ".wav",
+    "audio/wave": ".wav",
+    "audio/x-wav": ".wav",
+    "audio/mpeg": ".mp3",
+    "audio/mp3": ".mp3",
+    "audio/ogg": ".ogg",
+    "audio/flac": ".flac",
+    "audio/aac": ".aac",
+    "audio/m4a": ".m4a",
+    "audio/webm": ".webm",
+    "audio/mp4": ".mp4",
+    "video/mp4": ".mp4",
+    "video/webm": ".webm",
+}
+
+
+@component
+class FunASRTranscriber:
+    """
+    Transcribes audio files to Documents using [FunASR](https://github.com/modelscope/FunASR).
+
+    FunASR is an open-source speech recognition toolkit from Alibaba DAMO Academy.
+    It supports 50+ languages, speaker diarization, and timestamp extraction, and runs
+    entirely locally — no API key required.
+
+    Models are downloaded from ModelScope on first use and cached in `~/.cache/modelscope`.
+
+    **Usage Example:**
+
+    ```python
+    from haystack_integrations.components.audio.funasr import FunASRTranscriber
+
+    transcriber = FunASRTranscriber()
+    result = transcriber.run(sources=["speech.wav", "interview.mp3"])
+    documents = result["documents"]
+    ```
+
+    **Speaker diarization and punctuation:**
+
+    ```python
+    from haystack.utils import ComponentDevice
+
+    transcriber = FunASRTranscriber(
+        model="paraformer-zh",
+        vad_model="fsmn-vad",
+        punc_model="ct-punc",
+        spk_model="cam++",
+        device=ComponentDevice.from_str("cuda"),
+    )
+    ```
+
+    **SenseVoice with inverse text normalisation:**
+
+    ```python
+    transcriber = FunASRTranscriber(
+        model="iic/SenseVoiceSmall",
+        generation_kwargs={"use_itn": True, "merge_vad": True, "language": "auto"},
+    )
+    ```
+    """
+
+    def __init__(
+        self,
+        *,
+        model: str = "iic/SenseVoiceSmall",
+        vad_model: str | None = "fsmn-vad",
+        punc_model: str | None = "ct-punc",
+        spk_model: str | None = None,
+        device: ComponentDevice | None = None,
+        batch_size_s: int = 300,
+        store_full_path: bool = False,
+        generation_kwargs: dict[str, Any] | None = None,
+    ) -> None:
+        """
+        Create a FunASRTranscriber component.
+
+        :param model: FunASR model name or local path. Defaults to `"iic/SenseVoiceSmall"`,
+            a multilingual model supporting 50+ languages that is 5-10x faster than Whisper.
+            Alternatives include `"paraformer-zh"` (Chinese) or `"paraformer-en"` (English).
+            Browse available models at https://modelscope.github.io/FunASR/model-selection.html.
+        :param vad_model: Voice activity detection model used to split long audio into segments.
+            Set to `None` to process the audio as a single stream.
+            Browse available VAD models at https://www.modelscope.cn/models.
+        :param punc_model: Punctuation restoration model. Set to `None` to disable punctuation.
+            Browse available punctuation models at https://www.modelscope.cn/models.
+        :param spk_model: Speaker diarization model (e.g. `"cam++"`). When set, a `"speakers"`
+            key is included in the Document metadata. Defaults to `None` (diarization disabled).
+            Browse available speaker diarization models at https://www.modelscope.cn/models.
+        :param device: The device to run inference on. If `None`, the default device is selected
+            automatically. Use `ComponentDevice.from_str("cuda")` for GPU inference.
+        :param batch_size_s: Batch size in seconds for VAD-segmented audio. Larger values
+            improve throughput at the cost of memory.
+        :param store_full_path: If `True`, store the full audio file path in Document metadata.
+            If `False` (default), store only the file name.
+        :param generation_kwargs: Extra keyword arguments forwarded to `AutoModel.generate()`.
+            Use this for model-specific options such as `use_itn=True` or `merge_vad=True`
+            for SenseVoice, or `hotword="..."` for contextual recognition.
+        """
+        self.model = model
+        self.vad_model = vad_model
+        self.punc_model = punc_model
+        self.spk_model = spk_model
+        self.device = device
+        self.batch_size_s = batch_size_s
+        self.store_full_path = store_full_path
+        self.generation_kwargs: dict[str, Any] = generation_kwargs or {}
+        self._asr_model: Any = None
+
+    def warm_up(self) -> None:
+        """
+        Load the FunASR model into memory.
+
+        Models are downloaded from ModelScope on first call and cached locally.
+        This method is idempotent — calling it multiple times is safe.
+        """
+        if self._asr_model is not None:
+            return
+        self._asr_model = AutoModel(
+            model=self.model,
+            vad_model=self.vad_model,
+            punc_model=self.punc_model,
+            spk_model=self.spk_model,
+            device=ComponentDevice.resolve_device(self.device).to_torch_str(),
+        )
+
+    def to_dict(self) -> dict[str, Any]:
+        """
+        Serialize the component to a dictionary.
+
+        :returns: Dictionary with serialized data.
+        """
+        return default_to_dict(
+            self,
+            model=self.model,
+            vad_model=self.vad_model,
+            punc_model=self.punc_model,
+            spk_model=self.spk_model,
+            device=self.device.to_dict() if self.device else None,
+            batch_size_s=self.batch_size_s,
+            store_full_path=self.store_full_path,
+            generation_kwargs=self.generation_kwargs or None,
+        )
+
+    @classmethod
+    def from_dict(cls, data: dict[str, Any]) -> "FunASRTranscriber":
+        """
+        Deserialize the component from a dictionary.
+
+        :param data: Dictionary to deserialize from.
+        :returns: Deserialized component.
+        """
+        init_params = data.get("init_parameters", {})
+        if init_params.get("device"):
+            init_params["device"] = ComponentDevice.from_dict(init_params["device"])
+        return default_from_dict(cls, data)
+
+    @component.output_types(documents=list[Document])
+    def run(
+        self,
+        sources: list[str | Path | ByteStream],
+        meta: dict[str, Any] | list[dict[str, Any]] | None = None,
+    ) -> dict[str, list[Document]]:
+        """
+        Transcribe audio sources to Documents.
+
+        :param sources: Audio file paths (`str` or `Path`) or `ByteStream` objects.
+            Supported formats: WAV, MP3, FLAC, OGG, M4A, AAC, and any format that
+            FunASR's underlying audio backend (soundfile/ffmpeg) can decode.
+        :param meta: Metadata to attach to the produced Documents. Pass a single dict
+            to apply the same metadata to all Documents, or a list aligned with `sources`.
+        :returns: Dictionary with key `"documents"` — one `Document` per source whose
+            `content` holds the full transcript text.
+        """
+        if not sources:
+            return {"documents": []}
+        self.warm_up()
+
+        meta_list = normalize_metadata(meta, sources_count=len(sources))
+        documents: list[Document] = []
+
+        for source, user_meta in zip(sources, meta_list, strict=True):
+            try:
+                doc = self._transcribe(source, user_meta)
+                if doc is not None:
+                    documents.append(doc)
+            except Exception as e:
+                logger.warning(
+                    "Could not transcribe {source}. Skipping it. Error: {error}",
+                    source=source,
+                    error=e,
+                )
+
+        return {"documents": documents}
+
+    def _transcribe(
+        self,
+        source: str | Path | ByteStream,
+        user_meta: dict[str, Any],
+    ) -> Document | None:
+        """Transcribe a single audio source and return a Document, or None on empty result."""
+        tmp_path: str | None = None
+
+        if isinstance(source, ByteStream):
+            suffix = _mime_to_suffix(source.mime_type)
+            with tempfile.NamedTemporaryFile(suffix=suffix, delete=False) as tmp:
+                tmp.write(source.data)
+                tmp_path = tmp.name
+            audio_path: str = tmp_path
+            raw_name: str = str(source.meta.get("file_path") or source.meta.get("file_name") or "audio")
+            source_name = raw_name if self.store_full_path else Path(raw_name).name
+        else:
+            audio_path = str(source)
+            source_name = str(source) if self.store_full_path else Path(source).name
+
+        try:
+            results: list[dict[str, Any]] = self._asr_model.generate(
+                input=audio_path,
+                batch_size_s=self.batch_size_s,
+                **self.generation_kwargs,
+            )
+
+            if not results:
+                logger.warning("FunASR returned no results for {source}.", source=source)
+                return None
+
+            text = " ".join(r.get("text", "") for r in results).strip()
+
+            doc_meta: dict[str, Any] = {"file_path": source_name}
+
+            timestamps = [ts for r in results for ts in (r.get("timestamp") or [])]
+            if timestamps:
+                doc_meta["timestamps"] = timestamps
+
+            speakers = [sp for r in results for sp in (r.get("spk") or [])]
+            if speakers:
+                doc_meta["speakers"] = speakers
+
+            doc_meta.update(user_meta)
+            return Document(content=text, meta=doc_meta)
+        finally:
+            if tmp_path is not None:
+                with contextlib.suppress(OSError):
+                    os.unlink(tmp_path)
+
+
+def _mime_to_suffix(mime_type: str | None) -> str:
+    """Return a file extension for the given MIME type, defaulting to `.wav`."""
+    if mime_type:
+        return _MIME_TO_SUFFIX.get(mime_type.lower(), ".wav")
+    return ".wav"

funasr_haystack-0.0.1/tests/__init__.py

@@ -0,0 +1,3 @@
+# SPDX-FileCopyrightText: 2022-present deepset GmbH <info@deepset.ai>
+#
+# SPDX-License-Identifier: Apache-2.0

funasr_haystack-0.0.1/tests/conftest.py

@@ -0,0 +1,8 @@
+from pathlib import Path
+
+import pytest
+
+
+@pytest.fixture()
+def test_files_path():
+    return Path(__file__).parent / "test_files"

funasr_haystack-0.0.1/tests/test_transcriber.py

@@ -0,0 +1,365 @@
+# SPDX-FileCopyrightText: 2023-present deepset GmbH <info@deepset.ai>
+#
+# SPDX-License-Identifier: Apache-2.0
+
+import os
+import sys
+from unittest.mock import MagicMock, patch
+
+import pytest
+from haystack.dataclasses import ByteStream
+from haystack.utils import ComponentDevice
+
+from haystack_integrations.components.audio.funasr import FunASRTranscriber
+
+
+def _make_transcriber(**kwargs) -> FunASRTranscriber:
+    return FunASRTranscriber(**kwargs)
+
+
+def _inject_mock_model(transcriber: FunASRTranscriber, results: list[dict]) -> MagicMock:
+    mock_model = MagicMock()
+    mock_model.generate.return_value = results
+    transcriber._asr_model = mock_model
+    return mock_model
+
+
+class TestFunASRTranscriberInit:
+    def test_defaults(self):
+        t = _make_transcriber()
+        assert t.model == "iic/SenseVoiceSmall"
+        assert t.vad_model == "fsmn-vad"
+        assert t.punc_model == "ct-punc"
+        assert t.spk_model is None
+        assert t.device is None
+        assert t.batch_size_s == 300
+        assert t.store_full_path is False
+        assert t.generation_kwargs == {}
+        assert t._asr_model is None
+
+    def test_custom_params(self):
+        t = _make_transcriber(
+            model="paraformer-zh",
+            vad_model=None,
+            punc_model=None,
+            spk_model="cam++",
+            device=ComponentDevice.from_str("cuda"),
+            batch_size_s=60,
+            store_full_path=True,
+            generation_kwargs={"use_itn": True},
+        )
+        assert t.model == "paraformer-zh"
+        assert t.vad_model is None
+        assert t.punc_model is None
+        assert t.spk_model == "cam++"
+        assert t.device == ComponentDevice.from_str("cuda")
+        assert t.batch_size_s == 60
+        assert t.store_full_path is True
+        assert t.generation_kwargs == {"use_itn": True}
+
+
+class TestFunASRTranscriberSerialization:
+    def test_to_dict(self):
+        t = _make_transcriber()
+        d = t.to_dict()
+        assert d["type"].endswith("FunASRTranscriber")
+        p = d["init_parameters"]
+        assert p["model"] == "iic/SenseVoiceSmall"
+        assert p["vad_model"] == "fsmn-vad"
+        assert p["punc_model"] == "ct-punc"
+        assert p["spk_model"] is None
+        assert p["device"] is None
+        assert p["batch_size_s"] == 300
+        assert p["store_full_path"] is False
+        assert p["generation_kwargs"] is None
+
+    def test_to_dict_with_generation_kwargs(self):
+        t = _make_transcriber(generation_kwargs={"use_itn": True, "merge_vad": True})
+        d = t.to_dict()
+        assert d["init_parameters"]["generation_kwargs"] == {"use_itn": True, "merge_vad": True}
+
+    def test_to_dict_with_device(self):
+        t = _make_transcriber(device=ComponentDevice.from_str("cuda"))
+        d = t.to_dict()
+        assert d["init_parameters"]["device"] == {"type": "single", "device": "cuda"}
+
+    def test_from_dict(self):
+        data = {
+            "type": "haystack_integrations.components.audio.funasr.transcriber.FunASRTranscriber",
+            "init_parameters": {
+                "model": "paraformer-zh",
+                "vad_model": "fsmn-vad",
+                "punc_model": "ct-punc",
+                "spk_model": "cam++",
+                "device": {"type": "single", "device": "cuda"},
+                "batch_size_s": 60,
+                "store_full_path": False,
+                "generation_kwargs": {"use_itn": True},
+            },
+        }
+        t = FunASRTranscriber.from_dict(data)
+        assert t.model == "paraformer-zh"
+        assert t.spk_model == "cam++"
+        assert t.device == ComponentDevice.from_str("cuda")
+        assert t.generation_kwargs == {"use_itn": True}
+
+    def test_from_dict_no_device(self):
+        data = {
+            "type": "haystack_integrations.components.audio.funasr.transcriber.FunASRTranscriber",
+            "init_parameters": {
+                "model": "iic/SenseVoiceSmall",
+                "vad_model": "fsmn-vad",
+                "punc_model": "ct-punc",
+                "spk_model": None,
+                "device": None,
+                "batch_size_s": 300,
+                "store_full_path": False,
+                "generation_kwargs": None,
+            },
+        }
+        t = FunASRTranscriber.from_dict(data)
+        assert t.device is None
+
+    def test_to_from_dict_roundtrip(self):
+        t = _make_transcriber(
+            model="paraformer-zh",
+            spk_model="cam++",
+            batch_size_s=60,
+            device=ComponentDevice.from_str("cpu"),
+        )
+        t2 = FunASRTranscriber.from_dict(t.to_dict())
+        assert t.model == t2.model
+        assert t.spk_model == t2.spk_model
+        assert t.batch_size_s == t2.batch_size_s
+        assert t.device == t2.device
+
+
+class TestFunASRTranscriberWarmUp:
+    def test_warm_up_loads_model(self):
+        t = _make_transcriber()
+        mock_instance = MagicMock()
+        mock_cls = MagicMock(return_value=mock_instance)
+
+        with patch("haystack_integrations.components.audio.funasr.transcriber.AutoModel", mock_cls):
+            t.warm_up()
+
+        assert t._asr_model is mock_instance
+        expected_device = ComponentDevice.resolve_device(None).to_torch_str()
+        mock_cls.assert_called_once_with(
+            model="iic/SenseVoiceSmall",
+            vad_model="fsmn-vad",
+            punc_model="ct-punc",
+            spk_model=None,
+            device=expected_device,
+        )
+
+    def test_warm_up_idempotent(self):
+        t = _make_transcriber()
+        existing_model = MagicMock()
+        t._asr_model = existing_model
+        t.warm_up()
+        assert t._asr_model is existing_model
+
+
+class TestFunASRTranscriberRun:
+    def test_run_empty_sources(self):
+        t = _make_transcriber()
+        result = t.run(sources=[])
+        assert result == {"documents": []}
+
+    def test_run_single_file(self, tmp_path):
+        t = _make_transcriber()
+        audio_file = tmp_path / "test.wav"
+        audio_file.write_bytes(b"fake audio")
+        _inject_mock_model(t, [{"text": "Hello world"}])
+
+        result = t.run(sources=[str(audio_file)])
+        assert len(result["documents"]) == 1
+        assert result["documents"][0].content == "Hello world"
+        assert result["documents"][0].meta["file_path"] == "test.wav"
+
+    def test_run_multiple_files(self, tmp_path):
+        t = _make_transcriber()
+        f1, f2 = tmp_path / "a.wav", tmp_path / "b.wav"
+        f1.write_bytes(b"x")
+        f2.write_bytes(b"x")
+        mock = _inject_mock_model(t, [{"text": "one"}])
+        mock.generate.side_effect = [
+            [{"text": "one"}],
+            [{"text": "two"}],
+        ]
+
+        result = t.run(sources=[str(f1), str(f2)])
+        assert len(result["documents"]) == 2
+        assert result["documents"][0].content == "one"
+        assert result["documents"][1].content == "two"
+
+    def test_run_merges_vad_segments(self, tmp_path):
+        t = _make_transcriber()
+        audio = tmp_path / "long.wav"
+        audio.write_bytes(b"x")
+        _inject_mock_model(t, [{"text": "Hello"}, {"text": "world"}])
+
+        result = t.run(sources=[str(audio)])
+        assert result["documents"][0].content == "Hello world"
+
+    def test_run_with_metadata(self, tmp_path):
+        t = _make_transcriber()
+        audio = tmp_path / "test.wav"
+        audio.write_bytes(b"x")
+        _inject_mock_model(t, [{"text": "Hi"}])
+
+        result = t.run(sources=[str(audio)], meta={"topic": "greeting"})
+        assert result["documents"][0].meta["topic"] == "greeting"
+
+    def test_run_with_metadata_list(self, tmp_path):
+        t = _make_transcriber()
+        f1, f2 = tmp_path / "a.wav", tmp_path / "b.wav"
+        f1.write_bytes(b"x")
+        f2.write_bytes(b"x")
+        mock = _inject_mock_model(t, [{"text": "one"}])
+        mock.generate.side_effect = [[{"text": "one"}], [{"text": "two"}]]
+
+        result = t.run(sources=[str(f1), str(f2)], meta=[{"idx": 0}, {"idx": 1}])
+        assert result["documents"][0].meta["idx"] == 0
+        assert result["documents"][1].meta["idx"] == 1
+
+    def test_run_store_full_path(self, tmp_path):
+        t = _make_transcriber(store_full_path=True)
+        audio = tmp_path / "test.wav"
+        audio.write_bytes(b"x")
+        _inject_mock_model(t, [{"text": "Hi"}])
+
+        result = t.run(sources=[str(audio)])
+        assert result["documents"][0].meta["file_path"] == str(audio)
+
+    def test_run_with_timestamps(self, tmp_path):
+        t = _make_transcriber()
+        audio = tmp_path / "test.wav"
+        audio.write_bytes(b"x")
+        _inject_mock_model(t, [{"text": "Hi", "timestamp": [[0, 500], [500, 1000]]}])
+
+        result = t.run(sources=[str(audio)])
+        assert result["documents"][0].meta["timestamps"] == [[0, 500], [500, 1000]]
+
+    def test_run_with_speakers(self, tmp_path):
+        t = _make_transcriber(spk_model="cam++")
+        audio = tmp_path / "test.wav"
+        audio.write_bytes(b"x")
+        _inject_mock_model(t, [{"text": "Hi", "spk": [0, 1]}])
+
+        result = t.run(sources=[str(audio)])
+        assert result["documents"][0].meta["speakers"] == [0, 1]
+
+    def test_run_no_timestamps_or_speakers_when_absent(self, tmp_path):
+        t = _make_transcriber()
+        audio = tmp_path / "test.wav"
+        audio.write_bytes(b"x")
+        _inject_mock_model(t, [{"text": "Hi"}])
+
+        result = t.run(sources=[str(audio)])
+        assert "timestamps" not in result["documents"][0].meta
+        assert "speakers" not in result["documents"][0].meta
+
+    def test_run_empty_result_skipped(self, tmp_path):
+        t = _make_transcriber()
+        audio = tmp_path / "test.wav"
+        audio.write_bytes(b"x")
+        _inject_mock_model(t, [])
+
+        result = t.run(sources=[str(audio)])
+        assert result["documents"] == []
+
+    def test_run_error_skips_source(self, tmp_path):
+        t = _make_transcriber()
+        audio = tmp_path / "test.wav"
+        audio.write_bytes(b"x")
+        mock = _inject_mock_model(t, [])
+        mock.generate.side_effect = RuntimeError("model error")
+
+        result = t.run(sources=[str(audio)])
+        assert result["documents"] == []
+
+    def test_run_passes_generation_kwargs(self, tmp_path):
+        t = _make_transcriber(generation_kwargs={"use_itn": True, "language": "en"})
+        audio = tmp_path / "test.wav"
+        audio.write_bytes(b"x")
+        mock = _inject_mock_model(t, [{"text": "Hello"}])
+
+        t.run(sources=[str(audio)])
+        call_kwargs = mock.generate.call_args
+        assert call_kwargs.kwargs.get("use_itn") is True
+        assert call_kwargs.kwargs.get("language") == "en"
+        assert call_kwargs.kwargs.get("batch_size_s") == 300
+
+
+class TestFunASRTranscriberByteStream:
+    def test_run_bytestream(self):
+        t = _make_transcriber()
+        bs = ByteStream(data=b"fake audio", mime_type="audio/wav", meta={"file_name": "clip.wav"})
+        _inject_mock_model(t, [{"text": "Hello from bytes"}])
+
+        result = t.run(sources=[bs])
+        assert len(result["documents"]) == 1
+        assert result["documents"][0].content == "Hello from bytes"
+        assert result["documents"][0].meta["file_path"] == "clip.wav"
+
+    def test_run_bytestream_unknown_mime(self):
+        t = _make_transcriber()
+        bs = ByteStream(data=b"fake", mime_type=None)
+        mock = _inject_mock_model(t, [{"text": "text"}])
+
+        t.run(sources=[bs])
+        mock.generate.assert_called_once()
+
+    def test_run_bytestream_cleans_up_temp_file(self):
+        t = _make_transcriber()
+        bs = ByteStream(data=b"x", mime_type="audio/wav")
+        _inject_mock_model(t, [{"text": "hi"}])
+
+        created_paths: list[str] = []
+        original_ntf = __import__("tempfile").NamedTemporaryFile
+
+        def tracking_ntf(**kwargs):
+            f = original_ntf(**kwargs)
+            created_paths.append(f.name)
+            return f
+
+        with patch("tempfile.NamedTemporaryFile", side_effect=tracking_ntf):
+            t.run(sources=[bs])
+
+        for p in created_paths:
+            assert not os.path.exists(p), f"Temp file {p} was not cleaned up"
+
+    def test_run_bytestream_cleans_up_on_error(self):
+        t = _make_transcriber()
+        bs = ByteStream(data=b"x", mime_type="audio/wav")
+        mock = _inject_mock_model(t, [])
+        mock.generate.side_effect = RuntimeError("boom")
+
+        created_paths: list[str] = []
+        original_ntf = __import__("tempfile").NamedTemporaryFile
+
+        def tracking_ntf(**kwargs):
+            f = original_ntf(**kwargs)
+            created_paths.append(f.name)
+            return f
+
+        with patch("tempfile.NamedTemporaryFile", side_effect=tracking_ntf):
+            t.run(sources=[bs])
+
+        for p in created_paths:
+            assert not os.path.exists(p), f"Temp file {p} leaked after error"
+
+
+@pytest.mark.integration
+@pytest.mark.skipif(sys.version_info < (3, 14) or sys.platform != "linux", reason="slow")
+class TestFunASRTranscriberIntegration:
+    def test_transcribe_wav(self, test_files_path):
+        audio_path = test_files_path / "answer.wav"
+
+        t = FunASRTranscriber(model="iic/SenseVoiceSmall", vad_model=None, punc_model=None)
+        t.warm_up()
+        result = t.run(sources=[str(audio_path)])
+        assert len(result["documents"]) == 1
+        assert "answer" in result["documents"][0].content.lower()