update path to relative

scripts/asr_utils.py

@@ -4,7 +4,7 @@ import wave
 import re
 
 def add_text_index():
-    text_file = '/Users/jeqin/work/code/TestTranslator/tests/test_data/text/test_asr_zh.txt'
+    text_file = '../tests/test_data/text/test_asr_zh.txt'
     index = 1
     with open(text_file, encoding='utf-8') as f:
         for line in f:
@@ -45,9 +45,9 @@ def write_csv(rows, output_csv):
         writer.writerows(rows)
 
 def print_text_and_audio_length():
-    text_file = '/Users/jeqin/work/code/TestTranslator/tests/test_data/text/test_asr_zh_with_index.txt'
-    audio_folder = '/Users/jeqin/work/code/TestTranslator/tests/test_data/recordings'
-    output_csv = '/Users/jeqin/work/code/TestTranslator/scripts/csv/text_audio_length.csv'
+    text_file = '../tests/test_data/text/test_asr_zh_with_index.txt'
+    audio_folder = '../tests/test_data/recordings'
+    output_csv = 'csv/text_audio_length.csv'
     rows = []
     for idx, text in get_lines_with_index(text_file):
         # print(idx)
@@ -68,7 +68,7 @@ def get_text_distance(text1, text2):
     return d, nd, diff
 
 def get_origin_text_dict():
-    text_file = '/Users/jeqin/work/code/TestTranslator/tests/test_data/text/test_asr_zh_with_index.txt'
+    text_file = '../tests/test_data/text/test_asr_zh_with_index.txt'
     text_dict = {}
     for idx, text in get_lines_with_index(text_file):
         text_dict[idx] = text
@@ -77,5 +77,5 @@ def get_origin_text_dict():
 
 if __name__ == '__main__':
     # add_text_index()
-    # print_text_and_audio_length()
-    pass
+    print_text_and_audio_length()
+    # pass

scripts/run_funasr_quant.py

@@ -49,7 +49,7 @@ def inference(vad_model, asr_model, punc_model, audio:Path):
 def run_recordings():
     quantize = True
     vad_model, asr_model, punc_model = load_model(quantize)
-    audios = Path("/Users/jeqin/work/code/TestTranslator/tests/test_data/recordings/")
+    audios = Path("../tests/test_data/recordings/")
     rows = [["file_name", "time", "inference_result"]]
     original = get_origin_text_dict()
     for audio in sorted(audios.glob("*.wav"), key=lambda x: int(x.stem)):
@@ -62,7 +62,7 @@ def run_recordings():
 def run_test_audios():
     quantize = True
     vad_model, asr_model, punc_model = load_model(quantize)
-    audios = Path("/Users/jeqin/work/code/TestTranslator/tests/test_data/test_audios/")
+    audios = Path("../tests/test_data/test_audios/")
     rows = [["file_name", "time", "inference_result"]]
     for audio in sorted(audios.glob("*s/zh*.wav")):
         text, t = inference(vad_model, asr_model, punc_model, audio)

scripts/run_whisper.py

@@ -32,7 +32,7 @@ def load_model():
 
 def run_recordings():
     model = load_model()
-    audios = Path("/Users/jeqin/work/code/TestTranslator/tests/test_data/recordings/")
+    audios = Path("../tests/test_data/recordings/")
     rows = [["file_name", "time", "inference_result"]]
     original = get_origin_text_dict()
     for audio in sorted(audios.glob("*.wav"), key=lambda x: int(x.stem)):
@@ -51,7 +51,7 @@ def run_recordings():
 def run_test_audios():
     model = load_model()
     lang = "zh"
-    audios = Path("/Users/jeqin/work/code/TestTranslator/tests/test_data/test_audios/")
+    audios = Path("../tests/test_data/test_audios/")
     rows = [["file_name", "time", "inference_result"]]
     for audio in sorted(audios.glob(f"*{lang}*/*.wav")):
         print(audio)

scripts/{infer_finetuned_whisper.py → run_whisper_finetuned.py}

@@ -139,7 +139,7 @@ def load_model():
 
 def run_test_audios():
     model, processor = load_model()
-    audios = Path("/Users/jeqin/work/code/TestTranslator/tests/test_data/test_audios/")
+    audios = Path("../tests/test_data/test_audios/")
     rows = [["file_name", "inference_time", "inference_result"]]
     for audio in sorted(audios.glob("*en-ac1-16k/*.wav")): # *s/randomforest*.wav"
         try:
@@ -158,7 +158,7 @@ def run_test_audios():
 def run_recordings():
     from scripts.asr_utils import get_origin_text_dict, get_text_distance
     model, processor = load_model()
-    audios = Path("/Users/jeqin/work/code/TestTranslator/tests/test_data/recordings/")
+    audios = Path("../tests/test_data/recordings/")
     rows = [["file_name", "time", "inference_result"]]
     original = get_origin_text_dict()
     for audio in sorted(audios.glob("*.wav"), key=lambda x: int(x.stem)):