Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.
Python asraudiochineseemotion-recognitionfunasrmcp-servermultilingual-asropenai-compatible-apiparaformerpunctuationpytorchreal-time-asrspeaker-diarizationspeech-recognitionspeech-to-textstreaming-asrtranscriptionvllmvoice-activity-detectionwhisper-alternative
fix(runtime): release Torch CMVN parser fix as 0.1.6
zhifu gao9132d7881c78
diff --git a/runtime/python/libtorch/README.md b/runtime/python/libtorch/README.mdindex f06d374b0..00808b6ba 100644--- a/runtime/python/libtorch/README.md+++ b/runtime/python/libtorch/README.md@@ -27,12 +27,17 @@ includes the Contextual/Seaco Paraformer hotword batching and emptyhotword-file handling fix (#3838). This is a runtime wrapper fix, not amodel-weight update or an ASR accuracy claim.+Version 0.1.6 makes the standalone CMVN reader accept blank lines and+raise a file-specific error for missing or truncated statistics (#3842).+Valid float64 statistics and shipped feature output are unchanged. This is+a CPU parser/package fix, not a model-weight or ASR accuracy update.+install from pip```shell-pip install -U "funasr_torch>=0.1.5"+pip install -U "funasr_torch>=0.1.6"# For the users in China, you could install with the command:-# pip install -U "funasr_torch>=0.1.5" -i https://mirror.sjtu.edu.cn/pypi/web/simple+# pip install -U "funasr_torch>=0.1.6" -i https://mirror.sjtu.edu.cn/pypi/web/simple```or install from source codediff --git a/runtime/python/libtorch/setup.py b/runtime/python/libtorch/setup.pyindex 9b5607f69..be4b7e3c5 100644--- a/runtime/python/libtorch/setup.py+++ b/runtime/python/libtorch/setup.py@@ -15,7 +15,7 @@ def get_readme():setuptools.setup(name="funasr_torch",- version="0.1.5",+ version="0.1.6",platforms="Any",url="https://github.com/modelscope/FunASR.git",author="Speech Lab of DAMO Academy, Alibaba Group",diff --git a/tests/test_funasr_torch_package.py b/tests/test_funasr_torch_package.pyindex 4e25192e6..04cdcf01f 100644--- a/tests/test_funasr_torch_package.py+++ b/tests/test_funasr_torch_package.py@@ -119,12 +119,12 @@ def test_torch_package_does_not_require_unrelated_runtime_engines(distributions)}assert not names & {"torch-quant", "onnxruntime", "onnxruntime-gpu"}-def test_distribution_version_includes_hotword_batching_fix(distributions):+def test_distribution_version_includes_cmvn_parser_fix(distributions):from packaging.version import Versionwheel, sdist = distributionsversion = metadata(wheel)["Version"]- assert Version(version) >= Version("0.1.5")+ assert Version(version) >= Version("0.1.6")with tarfile.open(sdist, "r:gz") as archive:member = next(item for item in archive.getmembers()if item.name.count("/") == 1 and item.name.endswith("/PKG-INFO"))