From 465daa43fdbe1bd3479381f53e8e1fb878455932 Mon Sep 17 00:00:00 2001
From: zhifu gao <zhifu.gzf@alibaba-inc.com>
Date: 星期四, 16 二月 2023 10:12:00 +0800
Subject: [PATCH] Merge pull request #119 from alibaba-damo-academy/dev_jy
---
funasr/runtime/python/onnxruntime/paraformer/rapid_paraformer/paraformer_onnx.py | 17 +++++++++--------
1 files changed, 9 insertions(+), 8 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 f8fe063..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.utils import (CharTokenizer, Hypothesis, ONNXRuntimeError,
- OrtInferSession, TokenIDConverter, WavFrontend, get_logger,
+from utils.utils import (CharTokenizer, Hypothesis, ONNXRuntimeError,
+ OrtInferSession, TokenIDConverter, get_logger,
read_yaml)
-from .utils.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]
--
Gitblit v1.9.1