From cec84de3b676b5fdcd6c2f5dc30fe4b3571ed574 Mon Sep 17 00:00:00 2001
From: 游雁 <zhifu.gzf@alibaba-inc.com>
Date: 星期三, 15 二月 2023 20:09:21 +0800
Subject: [PATCH] Merge branch 'main' of github.com:alibaba-damo-academy/FunASR add
---
funasr/runtime/python/onnxruntime/paraformer/rapid_paraformer/paraformer_onnx.py | 26 +++++++++-----------------
1 files changed, 9 insertions(+), 17 deletions(-)
diff --git a/funasr/runtime/python/onnxruntime/paraformer/rapid_paraformer/paraformer_onnx.py b/funasr/runtime/python/onnxruntime/paraformer/rapid_paraformer/paraformer_onnx.py
index ca3d566..621b903 100644
--- a/funasr/runtime/python/onnxruntime/paraformer/rapid_paraformer/paraformer_onnx.py
+++ b/funasr/runtime/python/onnxruntime/paraformer/rapid_paraformer/paraformer_onnx.py
@@ -9,10 +9,11 @@
import librosa
import numpy as np
-from .utils import (CharTokenizer, Hypothesis, ONNXRuntimeError,
- OrtInferSession, TokenIDConverter, WavFrontend, get_logger,
+from utils.utils import (CharTokenizer, Hypothesis, ONNXRuntimeError,
+ OrtInferSession, TokenIDConverter, get_logger,
read_yaml)
-from .postprocess_utils import sentence_postprocess
+from utils.postprocess_utils import sentence_postprocess
+from utils.frontend import WavFrontend
logging = get_logger()
@@ -40,8 +41,8 @@
self.ort_infer = OrtInferSession(model_file, device_id)
self.batch_size = batch_size
- def __call__(self, wav_content: Union[str, np.ndarray, List[str]]) -> List:
- waveform_list = self.load_data(wav_content)
+ def __call__(self, wav_content: Union[str, np.ndarray, List[str]], fs: int = None) -> List:
+ waveform_list = self.load_data(wav_content, fs)
waveform_nums = len(waveform_list)
asr_res = []
@@ -62,10 +63,10 @@
return asr_res
def load_data(self,
- wav_content: Union[str, np.ndarray, List[str]]) -> List:
+ wav_content: Union[str, np.ndarray, List[str]], fs: int = None) -> List:
def load_wav(path: str) -> np.ndarray:
- waveform, _ = librosa.load(path, sr=None)
- return waveform[None, ...]
+ waveform, _ = librosa.load(path, sr=fs)
+ return waveform
if isinstance(wav_content, np.ndarray):
return [wav_content]
@@ -139,13 +140,4 @@
# text = self.tokenizer.tokens2text(token)
return text
-
-if __name__ == '__main__':
- project_dir = Path(__file__).resolve().parent.parent
- model_dir = "/home/zhifu.gzf/.cache/modelscope/hub/damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch"
- model = Paraformer(model_dir)
-
- wav_file = os.path.join(model_dir, 'example/asr_example.wav')
- result = model(wav_file)
- print(result)
--
Gitblit v1.9.1