diff --git a/environment.py b/environment.py
index a6d114b9959f654b52207135a75e7af42a63210f..14f754f81f787a13e67b86b445a629f653ccf467 100644
--- a/environment.py
+++ b/environment.py
@@ -26,4 +26,4 @@ class RunType(Enum):
code = 0
electron = 1
dev = 2
-RUN_TYPE = RunType.dev # electron or web
\ No newline at end of file
+RUN_TYPE = RunType.electron # electron or web
\ No newline at end of file
diff --git a/scripts/audios.txt b/scripts/audios.txt
new file mode 100644
index 0000000000000000000000000000000000000000..5c888ad390fbb57bebb467b5464efbce64acede0
--- /dev/null
+++ b/scripts/audios.txt
@@ -0,0 +1,70 @@
+/Users/jeqin/work/code/TestTranslator/tests/test_data/test_audios/5s-en-ac1-16k/English-chaos-part2-0.wav
+/Users/jeqin/work/code/TestTranslator/tests/test_data/test_audios/5s-en-ac1-16k/English-chaos-part2-5.wav
+/Users/jeqin/work/code/TestTranslator/tests/test_data/test_audios/5s-en-ac1-16k/English-chaos-part2-10.wav
+/Users/jeqin/work/code/TestTranslator/tests/test_data/test_audios/5s-en-ac1-16k/English-chaos-part2-15.wav
+/Users/jeqin/work/code/TestTranslator/tests/test_data/test_audios/10s-en-ac1-16k/English-chaos-part2-0.wav
+/Users/jeqin/work/code/TestTranslator/tests/test_data/test_audios/10s-en-ac1-16k/English-chaos-part2-10.wav
+/Users/jeqin/work/code/TestTranslator/tests/test_data/test_audios/10s-en-ac1-16k/English-chaos-part2-20.wav
+/Users/jeqin/work/code/TestTranslator/tests/test_data/test_audios/10s-en-ac1-16k/English-chaos-part2-30.wav
+/Users/jeqin/work/code/TestTranslator/tests/test_data/test_audios/English-chaos-part2.wav
+
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/es-1-0.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/es-1-5.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/es-1-10.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/es-1-15.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/es-1-0.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/es-1-10.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/es-1-20.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/es-1-30.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/es-1.wav
+
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/fr-1-0.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/fr-1-5.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/fr-1-10.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/fr-1-15.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/fr-1-0.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/fr-1-10.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/fr-1-20.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/fr-1-30.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/fr-1.wav
+
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/hi-2-0.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/hi-2-5.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/hi-2-10.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/hi-2-15.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/hi-2-0.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/hi-2-10.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/hi-2-20.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/hi-2-30.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/hi-2.wav
+
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/it-1-0.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/it-1-5.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/it-1-10.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/it-1-15.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/it-1-0.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/it-1-10.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/it-1-20.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/it-1-30.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/it-1.wav
+
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/ja-1-0.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/ja-1-5.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/ja-1-10.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/ja-1-15.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/ja-1-0.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/ja-1-10.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/ja-1-20.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/ja-1-30.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/ja-1.wav
+
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/pt-1-0.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/pt-1-5.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/pt-1-10.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/5s/pt-1-15.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/pt-1-0.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/pt-1-10.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/pt-1-20.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/10s/pt-1-30.wav
+/Users/jeqin/work/test/test_yoyotranslator/test_videos/audios/pt-1.wav
+
diff --git a/scripts/compare_whisper.py b/scripts/compare_whisper.py
new file mode 100644
index 0000000000000000000000000000000000000000..7fd1b91ef0f1d50afe4a53aa647b4f28cd5f7844
--- /dev/null
+++ b/scripts/compare_whisper.py
@@ -0,0 +1,92 @@
+from pywhispercpp.model import Model
+from pathlib import Path
+import time
+import csv
+
+from silero_vad.utils_vad import languages
+
+
+def save_csv(file_path, rows):
+ with open(file_path, "w", encoding="utf-8") as f:
+ writer = csv.writer(f)
+ writer.writerows(rows)
+ print(f"write csv to {file_path}")
+
+def run_audios_after_vad(models_dir, audio_dir, model_name):
+ t0 = time.time()
+ model = Model(
+ model=model_name,
+ models_dir=models_dir,
+ print_realtime=False,
+ print_progress=False,
+ print_timestamps=False,
+ translate=False,
+ # beam_search=1,
+ temperature=0.,
+ no_context=True
+ )
+ print("load model time: ", time.time()-t0)
+ rows = [["lang", "file_name", "inference_time", "python_res", "intel_res"]]
+ for lang in ["es", "fr", "hi", "it", "ja", "pt"]:
+ print("*" * 10, lang, "*"*10)
+ for audio in sorted(list((audio_dir/lang).glob("*.wav"))):
+ print("Audio name:", audio.name)
+ t1 = time.time()
+ output = model.transcribe(str(audio), language=lang)
+ t = time.time() - t1
+ print("Inference time:", t)
+ # print(output)
+ text = " ".join([a.text for a in output])
+ print("Text from Python:", text)
+ try:
+ with open(audio.with_suffix(".txt"), encoding="utf-8") as f:
+ intel_text = f.read().strip()
+ except Exception as e:
+ intel_text = ""
+ print(f"Error reading Intel text for {audio.name}: {e}")
+ print("Text from Intel :", intel_text)
+ rows.append([lang, audio.name, t, text, intel_text])
+ save_csv("csv/compare_whisper_intel.csv", rows)
+
+def run_long_audios(models_dir, audios_list, model_name):
+ t0 = time.time()
+ model = Model(
+ model=model_name,
+ models_dir=models_dir,
+ print_realtime=False,
+ print_progress=False,
+ print_timestamps=False,
+ translate=False,
+ # beam_search=1,
+ temperature=0.,
+ no_context=True
+ )
+ print("load model time: ", time.time() - t0)
+ rows = [["file_name", "inference_time", "res_text"]]
+ audios = audios_list.read_text().splitlines()
+ for audio in audios:
+ if not audio:
+ rows.append([])
+ continue
+ lang = Path(audio).name.split('-')[0]
+ if lang not in ["es", "fr", "hi", "it", "ja", "pt"]:
+ lang = "en"
+ print(f"Audio file: {audio}, lang: {lang}")
+ t1 = time.time()
+ output = model.transcribe(str(audio), language=lang)
+ t = time.time() - t1
+ print("Inference time:", t)
+ # print(output)
+ text = " ".join([a.text for a in output])
+ print("Text:", text)
+ rows.append([audio, t, text])
+ save_csv("csv/compare_whisper.csv", rows)
+if __name__ == '__main__':
+ models_dir = Path("/Users/jeqin/work/code/Translator/python_server/moyoyo_asr_models")
+ # model_name = "large-v3-turbo-q5_0"
+ model_name = "large-v3-turbo-q8_0"
+ # model_name = "small-q8_0"
+ # audios_after_vad = Path("/Users/jeqin/work/test/test_yoyotranslator/audios_after_vad/audio2-with-noise")
+ audios_list = Path("/Users/jeqin/work/code/TestTranslator/scripts/audios.txt")
+ # run_audios_after_vad(models_dir, audios_after_vad, model_name)
+ run_long_audios(models_dir, audios_list, model_name)
\ No newline at end of file
diff --git a/scripts/export_onnx.py b/scripts/export_onnx.py
new file mode 100644
index 0000000000000000000000000000000000000000..575f61ea3106fbf909c24933e852ca5b779c79d9
--- /dev/null
+++ b/scripts/export_onnx.py
@@ -0,0 +1,18 @@
+from funasr import AutoModel
+
+model_dir = "/Users/moyoyo/code/Translator/moyoyo_asr_models"
+asr_model_path = model_dir + '/speech_seaco_paraformer_large_asr_nat-zh-cn-16k-common-vocab8404-pytorch'
+vad_model_path = model_dir + '/speech_fsmn_vad_zh-cn-16k-common-pytorch'
+punc_model_path = model_dir + '/punc_ct-transformer_cn-en-common-vocab471067-large'
+
+model = AutoModel(model=asr_model_path)
+model_dir = model.export(type="onnx", quantize=True, disable_update=True)
+print(model_dir)
+
+model = AutoModel(model=vad_model_path)
+model_dir = model.export(type="onnx", quantize=True, disable_update=True)
+print(model_dir)
+
+model = AutoModel(model=punc_model_path)
+model_dir = model.export(type="onnx", quantize=True, disable_update=True)
+print(model_dir)
\ No newline at end of file
diff --git a/scripts/infer_finetuned_whisper.py b/scripts/infer_finetuned_whisper.py
new file mode 100644
index 0000000000000000000000000000000000000000..b43fe0c0eaf3ad341bae391cae3ffdff62879cea
--- /dev/null
+++ b/scripts/infer_finetuned_whisper.py
@@ -0,0 +1,157 @@
+import argparse
+import os
+import time
+from pathlib import Path
+import csv
+
+import numpy as np
+import torch
+import librosa
+from transformers import WhisperForConditionalGeneration, WhisperProcessor
+
+def save_csv(file_path, rows):
+ with open(file_path, "w", encoding="utf-8") as f:
+ writer = csv.writer(f)
+ writer.writerows(rows)
+ print(f"write csv to {file_path}")
+
+
+def load_audio(audio_path: str, sr: int = 16000):
+ # 读取音频并转成 16k 单声道 numpy float32
+ audio, _ = librosa.load(audio_path, sr=sr, mono=True)
+ return audio
+
+
+def transcribe_file(
+ audio_path: str,
+ model,
+ processor,
+ language: str = "Chinese",
+ task: str = "transcribe",
+ timestamps: bool = False,
+ max_new_tokens: int = 255,
+):
+ # 准备特征
+ audio = load_audio(audio_path, sr=16000)
+ inputs = processor(audio, sampling_rate=16000, return_tensors="pt")
+
+ # 放到设备
+ device = next(model.parameters()).device
+ input_features = inputs["input_features"].to(device)
+
+ # 生成
+ with torch.inference_mode(), torch.autocast(device_type="cuda", enabled=(device.type == "cuda")):
+ generated_ids = model.generate(
+ input_features=input_features,
+ max_new_tokens=max_new_tokens,
+ return_timestamps=timestamps, # 仅部分版本支持；不支持时自动忽略
+ )
+
+ # 解码
+ text = processor.tokenizer.batch_decode(generated_ids.cpu().numpy(), skip_special_tokens=True)
+ return text[0]
+
+
+def main():
+ parser = argparse.ArgumentParser("Simple Whisper Inference")
+ parser.add_argument("--model_path", type=str, default="whisper-large-v3-turbo-finetune",
+ help="本地合并模型路径或HF模型名")
+ parser.add_argument("--input", type=str, required=True,
+ help="音频文件路径，或目录（将批量处理其中的音频）")
+ parser.add_argument("--language", type=str, default="Chinese",
+ help="语言（如 Chinese / English / zh / en）")
+ parser.add_argument("--task", type=str, default="transcribe", choices=["transcribe", "translate"],
+ help="任务：转写或翻译")
+ parser.add_argument("--timestamps", action="store_true", help="是否返回时间戳（若模型与版本支持）")
+ parser.add_argument("--local_files_only", action="store_true", help="仅本地加载，不联网")
+ parser.add_argument("--batch_exts", type=str, default=".wav,.mp3,.flac,.m4a",
+ help="当 --input 是目录时，处理这些后缀的文件，逗号分隔")
+ args = parser.parse_args()
+
+ # 加载处理器 & 模型
+ processor = WhisperProcessor.from_pretrained(
+ args.model_path,
+ language=args.language,
+ task=args.task,
+ no_timestamps=not args.timestamps,
+ local_files_only=args.local_files_only,
+ )
+ model = WhisperForConditionalGeneration.from_pretrained(
+ args.model_path,
+ device_map="auto",
+ local_files_only=args.local_files_only,
+ torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32,
+ )
+
+ model.generation_config.language = args.language.lower()
+ model.generation_config.forced_decoder_ids = None
+ model.eval()
+
+ path = Path(args.input)
+ if path.is_file():
+ text = transcribe_file(
+ str(path), model, processor,
+ language=args.language, task=args.task, timestamps=args.timestamps
+ )
+ print(f"{path.name} -> {text}")
+ else:
+ # 目录批量
+ exts = {e.strip().lower() for e in args.batch_exts.split(",")}
+ files = [p for p in path.rglob("*") if p.suffix.lower() in exts]
+ if not files:
+ print("目录中未找到可处理的音频文件。")
+ return
+ for p in sorted(files):
+ try:
+ t0 = time.time()
+ text = transcribe_file(
+ str(p), model, processor,
+ language=args.language, task=args.task, timestamps=args.timestamps
+ )
+ t1 = time.time()
+ print(f"{p.name} -> {text}; time cost: {t1-t0}")
+ except Exception as e:
+ print(f"{p.name} -> 失败: {e}")
+def run():
+ model_path = "/Users/jeqin/Downloads/whisper-large-v3-turbo-finetune-0901"
+ lang = "en"
+ t0 = time.time()
+ processor = WhisperProcessor.from_pretrained(
+ model_path,
+ language=lang,
+ task="transcribe",
+ no_timestamps=True,
+ local_files_only=True,
+ )
+ model = WhisperForConditionalGeneration.from_pretrained(
+ model_path,
+ device_map="mps",
+ local_files_only=True,
+ torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32,
+ )
+
+ model.generation_config.language = lang.lower()
+ model.generation_config.forced_decoder_ids = None
+ model.eval()
+
+
+ print("load model time: ", time.time() - t0)
+ audios = Path("/Users/jeqin/work/code/TestTranslator/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:
+ t0 = time.time()
+ text = transcribe_file(
+ str(audio), model, processor
+ )
+
+ t = time.time()-t0
+ print(f"{audio.name} -> {text}; time cost: {t}")
+ rows.append([f"{audio.parent.name}/{audio.name}", t, text])
+ except Exception as e:
+ print(f"{audio.name} -> 失败: {e}")
+ save_csv("csv/fine-tune_whisper-0901.csv", rows)
+
+if __name__ == "__main__":
+ # main()
+ run()
diff --git a/scripts/run_funasr.py b/scripts/run_funasr.py
new file mode 100644
index 0000000000000000000000000000000000000000..f552077a228e2beeb9a50d4111164571758f0465
--- /dev/null
+++ b/scripts/run_funasr.py
@@ -0,0 +1,50 @@
+from funasr import AutoModel
+from pathlib import Path
+import time
+import csv
+
+def save_csv(file_path, rows):
+ with open(file_path, "w", encoding="utf-8") as f:
+ writer = csv.writer(f)
+ writer.writerows(rows)
+ print(f"write csv to {file_path}")
+
+def main():
+ model_dir = Path("/Users/jeqin/work/code/Translator/python_server/moyoyo_asr_models")
+
+ asr_model_path = model_dir / 'speech_seaco_paraformer_large_asr_nat-zh-cn-16k-common-vocab8404-pytorch'
+ vad_model_path = model_dir / 'speech_fsmn_vad_zh-cn-16k-common-pytorch'
+ punc_model_path = model_dir / 'punc_ct-transformer_cn-en-common-vocab471067-large'
+ t0 = time.time()
+ model = AutoModel(
+ model=asr_model_path.as_posix(),
+ vad_model=vad_model_path.as_posix(),
+ punc_model=punc_model_path.as_posix(),
+ log_level="ERROR",
+ disable_update=True
+ )
+ t1 = time.time()
+ print("load model: ", t1 - t0)
+ audios = Path("/Users/jeqin/work/code/TestTranslator/tests/test_data/test_audios/")
+ rows = [["file_name", "inference_time", "inference_result"]]
+ for audio in sorted(audios.glob("*ac1-16k/Chinese*")):
+ print(audio)
+ t1 = time.time()
+ try:
+ result = model.generate(input=str(audio), disable_pbar=True,
+ hotword="")
+ except Exception as e:
+ print(audio)
+ print(e)
+ t2 = time.time()
+ t = t2-t1
+ print("inference time:", t)
+ text = result[0]["text"]
+ print("inference result", text)
+ rows.append([f"{audio.parent.name}/{audio.name}", t, text])
+ save_csv(f"csv/funasr.csv", rows)
+
+
+
+if __name__ == '__main__':
+ main()
\ No newline at end of file
diff --git a/scripts/run_funasr_c.py b/scripts/run_funasr_c.py
new file mode 100644
index 0000000000000000000000000000000000000000..91f05002a9930733fb8722b1a337f3969942b92b
--- /dev/null
+++ b/scripts/run_funasr_c.py
@@ -0,0 +1,39 @@
+
+import sys
+import time
+from pathlib import Path
+import csv
+
+sys.path.append('/Users/jeqin/work/code/funasr_wrapper/build') # 添加编译生成的模块路径
+sys.path.append('/Users/jeqin/work/code/funasr_wrapper/build/src') # 添加编译生成的模块路径
+import funasr_py
+
+def save_csv(file_path, rows):
+ with open(file_path, "w", encoding="utf-8") as f:
+ writer = csv.writer(f)
+ writer.writerows(rows)
+ print(f"write csv to {file_path}")
+
+def main():
+ t0 = time.time()
+ config_file = "/Users/jeqin/work/code/funasr_wrapper/testpy/config.json"
+ asr = funasr_py.FunasrEasy(config_file)
+ # 初始化模型
+ asr.init()
+ t1 = time.time()
+ print("Initializing model: ", t1-t0)
+ audios = Path("/Users/jeqin/work/code/TestTranslator/tests/test_data/test_audios/")
+ rows = [["file_name", "inference_time", "inference_result"]]
+ for audio in sorted(audios.glob("*s-ac1/Chinese*")):
+ print(audio)
+ t1 = time.time()
+ result = asr.infer(str(audio))
+ text = asr.get_text(result)
+ asr.free_result(result)
+ t = time.time() - t1
+ print("inference time:", t)
+ print(text)
+ rows.append([f"{audio.parent.name}/{audio.name}", t, text])
+ save_csv("csv/funasr_c.csv", rows)
+if __name__ == '__main__':
+ main()
diff --git a/scripts/run_kokoro.py b/scripts/run_kokoro.py
new file mode 100644
index 0000000000000000000000000000000000000000..a452d7c931b3e834f713d94a2fed54e70cf2bd80
--- /dev/null
+++ b/scripts/run_kokoro.py
@@ -0,0 +1,54 @@
+from pathlib import Path
+from time import time
+import os
+
+import soundfile as sf
+from misaki import zh
+import onnxruntime
+
+from kokoro_onnx import Kokoro
+
+# providers = onnxruntime.get_available_providers()
+# print(f"Available onnx runtime providers: {providers}")
+
+def create_session(model_path):
+ # See list of providers https://github.com/microsoft/onnxruntime/issues/22101#issuecomment-2357667377
+ providers = onnxruntime.get_available_providers()
+ providers = providers[1:2]
+ print(f"Available onnx runtime providers: {providers}")
+
+
+ # See session options https://onnxruntime.ai/docs/performance/tune-performance/threading.html#thread-management
+ sess_options = onnxruntime.SessionOptions()
+ cpu_count = os.cpu_count() // 2
+ print(f"Setting threads to CPU cores count: {cpu_count}")
+ # sess_options.intra_op_num_threads = cpu_count
+ session = onnxruntime.InferenceSession(
+ model_path, providers=providers, sess_options=sess_options
+ )
+ return session
+
+model_folder = Path("/Users/jeqin/work/code/Translator/python_server/moyoyo_asr_models/kokoro")
+model_path = str(model_folder/"kokoro-quant.onnx")
+voice_model_path = str(model_folder/"voices-v1.0.bin")
+vocab_config = str(model_folder/"zh_config.json")
+
+texts = [
+ "千里之行，始于足下。",
+ "我想听你唱首歌",
+ "窗前明月光，疑是地上霜。举头望明月，低头思故乡。"
+]
+voice = "zf_xiaoyi"
+session = create_session(model_path)
+model = Kokoro.from_session(session, voice_model_path, vocab_config=vocab_config)
+g2p = zh.ZHG2P()
+for i in range(5):
+ for index, text in enumerate(texts):
+ phonemes, _ = g2p(text)
+ start = time()
+ samples, sample_rate = model.create(phonemes, voice=voice, speed=1.0, is_phonemes=True)
+ end = time()
+ time_cost = end - start
+ print(f"time cost: {time_cost} for text: {text}")
+ sf.write(f"audio_{index}.wav", samples, sample_rate)
+ print(f"Created audio_{index}.wav")
\ No newline at end of file
diff --git a/scripts/run_kokoro_sample.py b/scripts/run_kokoro_sample.py
new file mode 100644
index 0000000000000000000000000000000000000000..944dcc8ea78f5cacff199ba9426fbf0fe8940c4f
--- /dev/null
+++ b/scripts/run_kokoro_sample.py
@@ -0,0 +1,65 @@
+from time import time
+import soundfile as sf
+from misaki import en, espeak, zh
+from kokoro_onnx import Kokoro
+
+
+
+def run_en():
+ # Misaki G2P with espeak-ng fallback
+ fallback = espeak.EspeakFallback(british=False)
+ g2p = en.G2P(trf=False, british=False, fallback=fallback)
+
+ models = "/Users/jeqin/work/code/TestTranslator/scripts/kokoro_models/"
+ # Kokoro
+ kokoro = Kokoro(f"{models}kokoro-v1.0.onnx", f"{models}voices-v1.0.bin")
+
+ texts = [
+ "[Misaki](/misˈɑki/) is a G2P engine designed for [Kokoro](/kˈOkəɹO/) models.",
+ "For example, the geology and terrain along the railway line.",
+ " When choosing solid-state drives, we sometimes see reviews or videos discussing whether a particular solid-state drive has a caching scheme or an uncaching scheme in the performance testing section."
+ ]
+ for index, text in enumerate(texts):
+ # Phonemize
+ # text = "[Misaki](/misˈɑki/) is a G2P engine designed for [Kokoro](/kˈOkəɹO/) models."
+ phonemes, _ = g2p(text)
+
+ # Create
+ start = time()
+ samples, sample_rate = kokoro.create(phonemes, "af_heart", is_phonemes=True)
+ end = time()
+ time_cost = end - start
+ print(f"time cost: {time_cost} for text: {text}")
+ # Save
+ sf.write(f"audio{index}.wav", samples, sample_rate)
+ print(f"Created audio{index}.wav")
+
+def run_zh():
+ # Misaki G2P with espeak-ng fallback
+ # fallback = espeak.EspeakFallback(british=False)
+ g2p = zh.ZHG2P()
+
+ models = "/Users/jeqin/work/code/Translator/python_server/moyoyo_asr_models/kokoro"
+ # Kokoro
+ kokoro = Kokoro(f"{models}/kokoro-quant.onnx", f"{models}/voices-v1.0.bin", vocab_config=f"{models}/zh_config.json")
+
+ texts = [
+ "千里之行，始于足下。",
+ "我想听你唱首歌",
+ "窗前明月光，疑是地上霜。举头望明月，低头思故乡。"
+ ]
+ for index, text in enumerate(texts):
+ phonemes, _ = g2p(text)
+
+ # Create
+ start = time()
+ samples, sample_rate = kokoro.create(phonemes, "zf_xiaoyi", is_phonemes=True, speed=1.0)
+ end = time()
+ time_cost = end - start
+ print(f"time cost: {time_cost} for text: {text}")
+ # Save
+ sf.write(f"audio{index}.wav", samples, sample_rate)
+ print(f"Created audio{index}.wav")
+
+if __name__ == '__main__':
+ run_zh()
\ No newline at end of file
diff --git a/scripts/run_quant.py b/scripts/run_quant.py
new file mode 100644
index 0000000000000000000000000000000000000000..e82c9934369da54dbfc5eaba105832838cc6ebdd
--- /dev/null
+++ b/scripts/run_quant.py
@@ -0,0 +1,51 @@
+from pathlib import Path
+import time
+import csv
+from funasr_onnx import SeacoParaformer, CT_Transformer, Fsmn_vad
+
+def save_csv(file_path, rows):
+ with open(file_path, "w", encoding="utf-8") as f:
+ writer = csv.writer(f)
+ writer.writerows(rows)
+ print(f"write csv to {file_path}")
+def main():
+ model_dir = Path("/Users/jeqin/work/code/Translator/python_server/moyoyo_asr_models")
+
+ asr_model_path = model_dir / 'speech_seaco_paraformer_large_asr_nat-zh-cn-16k-common-vocab8404-pytorch'
+ vad_model_path = model_dir / 'speech_fsmn_vad_zh-cn-16k-common-pytorch'
+ punc_model_path = model_dir / 'punc_ct-transformer_cn-en-common-vocab471067-large'
+ t0 = time.time()
+ quantize = True
+ vad_model = Fsmn_vad(vad_model_path, quantize=quantize)
+ asr_model = SeacoParaformer(asr_model_path, quantize=quantize)
+ punc_model = CT_Transformer(punc_model_path, quantize=quantize)
+ t1 = time.time()
+ print("load model time:", t1 - t0)
+ audios = Path("/Users/jeqin/work/code/TestTranslator/tests/test_data/test_audios/")
+ rows = [["file_name", "inference_time", "inference_result"]]
+ for audio in sorted(audios.glob("*s/randomforest*.wav")):
+ t1 = time.time()
+ vad_res = vad_model(str(audio))
+ t2 = time.time()
+ print("vad time:", t2-t1)
+ asr_res = asr_model(str(audio), hotwords="")
+ asr_text = asr_res[0]["preds"]
+ t3 = time.time()
+ print("asr time:", t3-t2)
+ print("asr text:", asr_text)
+ result = punc_model(asr_text)
+ text = result[0]
+ t4 = time.time()
+ print("punc time:", t4-t3)
+ print("punc text:", text)
+ # print(text)
+ # vad_res = vad_model(str(audio))
+ # t5 = time.time()
+ # print("vad time:", t5 - t4)
+ t = t4-t1
+ print("inference:", t)
+ rows.append([f"{audio.parent.name}/{audio.name}", t, text])
+ file_name = "csv/quant.csv" if quantize else "run_onnx.csv"
+ save_csv(file_name, rows)
+if __name__ == '__main__':
+ main()
diff --git a/scripts/run_whisper.py b/scripts/run_whisper.py
index e35d4297ab203e8679a16074b99a8627485ea7e8..4a3736fe1bf318b53b54eab377b6ee45cdb1f741 100644
--- a/scripts/run_whisper.py
+++ b/scripts/run_whisper.py
@@ -1,26 +1,45 @@
from pywhispercpp.model import Model
from pathlib import Path
import time
+import csv
from silero_vad.utils_vad import languages
-models_dir = Path("/Users/jeqin/work/code/Translator/moyoyo_asr_models")
-whisper_model = 'large-v3-turbo-q5_0'
-model = Model(
- model=whisper_model,
- models_dir=models_dir,
- print_realtime=False,
- print_progress=False,
- print_timestamps=False,
- translate=False,
- # beam_search=1,
- temperature=0.,
- no_context=True
- )
-audios = Path("/Users/jeqin/work/code/TestTranslator/tests/test_data/test_audios")
-for audio in sorted(audios.glob("English*")):
- print(audio)
- t1 = time.time()
- output = model.transcribe(str(audio), language="en")
- print("inference time:", time.time()-t1)
- print(" ".join([a.text for a in output]))
\ No newline at end of file
+
+def save_csv(file_path, rows):
+ with open(file_path, "w", encoding="utf-8") as f:
+ writer = csv.writer(f)
+ writer.writerows(rows)
+ print(f"write csv to {file_path}")
+
+def main():
+ models_dir = Path("/Users/jeqin/work/code/Translator/python_server/moyoyo_asr_models")
+ whisper_model = 'large-v3-turbo-q5_0'
+ t0 = time.time()
+ model = Model(
+ model=whisper_model,
+ models_dir=models_dir,
+ print_realtime=False,
+ print_progress=False,
+ print_timestamps=False,
+ translate=False,
+ # beam_search=1,
+ temperature=0.,
+ no_context=True
+ )
+ print("load model time: ", time.time()-t0)
+ audios = Path("/Users/jeqin/work/code/TestTranslator/tests/test_data/test_audios/")
+ rows = [["file_name", "inference_time", "inference_result"]]
+ for audio in sorted(audios.glob("*-mix/randomforest*.wav")):
+ print(audio)
+ t1 = time.time()
+ output = model.transcribe(str(audio), language="zh")#, language="zh", initial_prompt="这是一段中文的会议内容。")# initial_prompt="这是一段中文的会议内容。"
+ t = time.time() - t1
+ print("inference time:", t)
+ text = " ".join([a.text for a in output])
+ print(text)
+ rows.append([f"{audio.parent.name}/{audio.name}", t, text])
+ # save_csv("csv/whisper.csv", rows)
+
+if __name__ == '__main__':
+ main()
\ No newline at end of file
diff --git a/scripts/split_audio.py b/scripts/split_audio.py
new file mode 100644
index 0000000000000000000000000000000000000000..00fb0508376ec4ecaad71772664d5a498f3120b7
--- /dev/null
+++ b/scripts/split_audio.py
@@ -0,0 +1,35 @@
+from pathlib import Path
+import subprocess
+from subprocess import CompletedProcess
+
+
+def cmd(command: str, check=True, capture_output=False) -> CompletedProcess:
+ print(command)
+ if capture_output:
+ ret = subprocess.run(command, shell=True, check=check, stdout=subprocess.PIPE, stderr=subprocess.STDOUT,
+ universal_newlines=True)
+ else:
+ ret = subprocess.run(command, shell=True, check=check)
+ print(ret.stdout)
+ return ret
+
+
+current = Path("/Users/jeqin/work/code/TestTranslator/tests/test_data/test_audios")
+audios_5s = current/"5s"
+audios_10s = current/"10s"
+if not audios_5s.exists():
+ audios_5s.mkdir(parents=True, exist_ok=True)
+if not audios_10s.exists():
+ audios_10s.mkdir(parents=True, exist_ok=True)
+for f in sorted(current.glob("randomforest*.wav")):
+ file_name = f.name
+ print(file_name)
+ for i in [0, 5, 10, 15]:
+ new_name = f"{f.stem}-{i}.wav"
+ # -ac 1 -ar 16000
+ command=f"ffmpeg -i {f} -ss 00:00:{str(i).zfill(2)} -ac 1 -ar 16000 -t 00:00:05 {audios_5s/new_name}"
+ cmd(command)
+ for i in [0, 10, 20, 30]:
+ new_name = f"{f.stem}-{i}.wav"
+ command = f"ffmpeg -i {f} -ss 00:00:{str(i).zfill(2)} -ac 1 -ar 16000 -t 00:00:10 {audios_10s/new_name}"
+ cmd(command)
diff --git a/temp.py b/temp.py
new file mode 100644
index 0000000000000000000000000000000000000000..bdf4888e7f4d5f0fa7f186a37452cc49d7261147
--- /dev/null
+++ b/temp.py
@@ -0,0 +1,4 @@
+text ="""
+{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0].role == 'system' %}\n {{- messages[0].content + '\\n\\n' }}\n {%- endif %}\n {{- \"# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within XML tags:\\n\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n\\n\\nFor each function call, return a json object with function name and arguments within XML tags:\\n\\n{\\\"name\\\": , \\\"arguments\\\": }\\n<|im_end|>\\n\" }}\n{%- else %}\n {%- if messages[0].role == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0].content + '<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n {%- set index = (messages|length - 1) - loop.index0 %}\n {%- if ns.multi_step_tool and message.role == \"user\" and message.content is string and not(message.content.startswith('') and message.content.endswith('')) %}\n {%- set ns.multi_step_tool = false %}\n {%- set ns.last_query_index = index %}\n {%- endif %}\n{%- endfor %}\n{%- for message in messages %}\n {%- if message.content is string %}\n {%- set content = message.content %}\n {%- else %}\n {%- set content = '' %}\n {%- endif %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {%- set reasoning_content = '' %}\n {%- if message.reasoning_content is string %}\n {%- set reasoning_content = message.reasoning_content %}\n {%- else %}\n {%- if '' in content %}\n {%- set reasoning_content = content.split('')[0].rstrip('\\n').split('')[-1].lstrip('\\n') %}\n {%- set content = content.split('')[-1].lstrip('\\n') %}\n {%- endif %}\n {%- endif %}\n {%- if loop.index0 > ns.last_query_index %}\n {%- if loop.last or (not loop.last and reasoning_content) %}\n {{- '<|im_start|>' + message.role + '\\n\\n' + reasoning_content.strip('\\n') + '\\n\\n\\n' + content.lstrip('\\n') }}\n {%- else %}\n {{- '<|im_start|>' + message.role + '\\n' + content }}\n {%- endif %}\n {%- else %}\n {{- '<|im_start|>' + message.role + '\\n' + content }}\n {%- endif %}\n {%- if message.tool_calls %}\n {%- for tool_call in message.tool_calls %}\n {%- if (loop.first and content) or (not loop.first) %}\n {{- '\\n' }}\n {%- endif %}\n {%- if tool_call.function %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {%- if tool_call.arguments is string %}\n {{- tool_call.arguments }}\n {%- else %}\n {{- tool_call.arguments | tojson }}\n {%- endif %}\n {{- '}\\n' }}\n {%- endfor %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n\\n' }}\n {{- content }}\n {{- '\\n' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}
+"""
+print(text)
\ No newline at end of file
diff --git a/tests/test_accuracy_and_delay.py b/tests/test_accuracy_and_delay.py
index daeb9a9568dd4aa537a432be1203b0d58d50982b..5708303a7934224ca0bdd6dc275aa3bb59cadb9e 100644
--- a/tests/test_accuracy_and_delay.py
+++ b/tests/test_accuracy_and_delay.py
@@ -14,7 +14,7 @@ def test_accuracy_and_delay_zh2en(app, log_file, page: TranslatorPage,
audio:Path,):
page.start_zh2en()
translation_lang = "zh2en"
- time.sleep(2)
+ time.sleep(3)
audio_length = get_length(audio)
play_audio(audio)
web_records = page.get_current_node_text(duration=audio_length)
@@ -38,7 +38,7 @@ def test_accuracy_and_delay_en2zh(app, log_file, page: TranslatorPage,
audio:Path):
page.start_en2zh()
translation_lang = "en2zh"
- time.sleep(2)
+ time.sleep(3)
audio_length = get_length(audio)
play_audio(audio)
web_records = page.get_current_node_text(duration=audio_length)
diff --git a/tests/test_data/test_audios.zip b/tests/test_data/test_audios.zip
new file mode 100644
index 0000000000000000000000000000000000000000..9051f7e3cae14fe1cda5b7eea1ffdb518d9b379c
--- /dev/null
+++ b/tests/test_data/test_audios.zip
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b3832d0c066ab144e2cda7e37df5144922dbceb0ae2605134eada3c866b0d43
+size 83025760
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-chaos-part2-0.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-chaos-part2-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..964a99099baad98ba8313d0ac7dc821cf1e020a6
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-chaos-part2-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:111f098aa42c139e19c795fc65b14d3b1435a29d75d208592c59e98f5e43144a
+size 320078
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-chaos-part2-10.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-chaos-part2-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..c4e36a18ae94e0b06e9b44baef87e2e7e6568773
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-chaos-part2-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:836b29596a0c1609aa91d6d48bc3fd7c73ebda89656744d5ba5691168bebc8a7
+size 320078
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-chaos-part2-20.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-chaos-part2-20.wav
new file mode 100644
index 0000000000000000000000000000000000000000..54524204f21a17d0d11e3a96265bb55ce33443b2
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-chaos-part2-20.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a2f7a0f2e768846ad361672b402bb243c30c22631286908f58a8ffb9d4361ad
+size 320078
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-chaos-part2-30.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-chaos-part2-30.wav
new file mode 100644
index 0000000000000000000000000000000000000000..74e8e918a8ddd51b8eb9af1486e6c6188b0ca3a4
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-chaos-part2-30.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef83616e756e449f0307c93b97b8d260bc4c68e213dc878e9f0ca4a46e2a69b7
+size 320078
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-computer_sicence-part1-0.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-computer_sicence-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..95e2a0630dd7f1c8851b8c295d875d47390b2b81
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-computer_sicence-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dbf8ee215f2e447dff00e5d3cfee257a2945f1689c5af2fd995729f02315802d
+size 320078
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-computer_sicence-part1-10.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-computer_sicence-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..6aad5834ba7278ef28abb1e7f4f947a7fe6de07c
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-computer_sicence-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e27e550cde48277d1239cb9d9ee40749c520c2f4d5824bfb2fb46b29a8db2fc8
+size 320078
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-computer_sicence-part1-20.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-computer_sicence-part1-20.wav
new file mode 100644
index 0000000000000000000000000000000000000000..8d4a7367ed0e5e1e54dd03cb6664daeeeeec9e6e
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-computer_sicence-part1-20.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69bedb0d844819919c44f5280aeb8ce20d3eee30099565bfff926aa883702a3c
+size 320078
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-computer_sicence-part1-30.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-computer_sicence-part1-30.wav
new file mode 100644
index 0000000000000000000000000000000000000000..baff72f4987b57f666384517fd9bfb7c6211ab79
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-computer_sicence-part1-30.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:606f1b5ad35f4edeaf274fd7e54c1b32cf22e905feee795c207de9d837f9031a
+size 320078
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-generative_ai-part1-0.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-generative_ai-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..9d8d936f28aaf1488a8ed9f9875edaeddf887cd5
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-generative_ai-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:171b9e830af34320b9049564390b36305fd98168a82a45d3bb93f24acb2ede29
+size 320078
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-generative_ai-part1-10.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-generative_ai-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..979ca5f03b9f7a25929ddc8c87d70a52baac6f3f
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-generative_ai-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7077bd545e60a791a4f5bef34e6d52a0c580ea7f3cb767bc2808442836347ec5
+size 320078
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-generative_ai-part1-20.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-generative_ai-part1-20.wav
new file mode 100644
index 0000000000000000000000000000000000000000..6c442fdaf28a0423eb600c3ce7c266a8979a3492
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-generative_ai-part1-20.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a85b17f66cc09d5cbdd55f17e52bc376db9d7c3668a1613b74cc60c146a8b2aa
+size 320078
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-generative_ai-part1-30.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-generative_ai-part1-30.wav
new file mode 100644
index 0000000000000000000000000000000000000000..f5d0265760cbebdd95b9582c4d0264d744eadf5e
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-generative_ai-part1-30.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b25e9278c64ba162139f4785a0419435112653d3c7bd66bb7d6a35e7c20bd12b
+size 320078
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-internet-part20-0.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-internet-part20-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..cb42daa339201955bdc1cb08380a35b2095f48f2
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-internet-part20-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2661796edf3667289ebb1772c3a3fb3d120ae7fb2e96c08899a5261b817fef49
+size 320078
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-internet-part20-10.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-internet-part20-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..e3ec1db84effb5e97e1c14627039bbfd6ecd188a
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-internet-part20-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:100272f678ee41dc71c35e1b705bfbf3aef69650562539a1390f87a8ec21a926
+size 320078
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-internet-part20-20.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-internet-part20-20.wav
new file mode 100644
index 0000000000000000000000000000000000000000..0d930b7f90de6a569656e400b5455ce004e0a519
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-internet-part20-20.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5110fd6774eab81a40c6c11fe5b08cb941588aa3fec0a00aa6bc951907750dec
+size 320078
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-internet-part20-30.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-internet-part20-30.wav
new file mode 100644
index 0000000000000000000000000000000000000000..2421192705d4619b3fc67e12441c16884f647af3
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-internet-part20-30.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ace6cb64e830ab31892ea4ca072051b5993bbe393ac63bb887157b6c6808bbf6
+size 320078
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-legalsystem-part1-0.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-legalsystem-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..ddbd177e43696218af30e15d52f2a3dcf692b91a
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-legalsystem-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b6d1ba99abc344ec4d31a3a6e34af5ab81dda139512f306bba94c5c52b71edc
+size 320078
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-legalsystem-part1-10.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-legalsystem-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..76566c206f8bb1b76eed48eab3a8e859888c7785
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-legalsystem-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48e5aff409e3b83ae387dd6fd9c06f131116191161e2a844ba69febf1e8cbf3f
+size 320078
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-legalsystem-part1-20.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-legalsystem-part1-20.wav
new file mode 100644
index 0000000000000000000000000000000000000000..dff506eeac3234df5f1f4c68e7a5902de510fe94
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-legalsystem-part1-20.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bfea77a95b17e3dace6dd5504bb5d618597619e1e64ecc91e52748e354331170
+size 320078
diff --git a/tests/test_data/test_audios/10s-en-ac1-16k/English-legalsystem-part1-30.wav b/tests/test_data/test_audios/10s-en-ac1-16k/English-legalsystem-part1-30.wav
new file mode 100644
index 0000000000000000000000000000000000000000..3c40ccc0d1078727299af18409e56e020cabb7f7
--- /dev/null
+++ b/tests/test_data/test_audios/10s-en-ac1-16k/English-legalsystem-part1-30.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0111f1d6d65692dce0b3200de36af5fa27b086401e462123baddc11f45fc8ef6
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/qiaodan-part1-0.wav b/tests/test_data/test_audios/10s-mix/qiaodan-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..d454acac301ebf2326dc51f4c41622b546961019
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/qiaodan-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:464420cd14b00b08d4240fb5ecc19b2aa053ebbf534cb0aadb8f80f7bf0da668
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/qiaodan-part1-10.wav b/tests/test_data/test_audios/10s-mix/qiaodan-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..1a0914141cda61f40059dbe82b388868331c592e
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/qiaodan-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5dbd856cf2a9543f3c133483708a6354a9ee06718b496bce03c29a981a56f8fc
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/qiaodan-part1-20.wav b/tests/test_data/test_audios/10s-mix/qiaodan-part1-20.wav
new file mode 100644
index 0000000000000000000000000000000000000000..188abdd7a4e2b58743f4091b1d019777e74a6bbe
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/qiaodan-part1-20.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:59c1d0e112e1fafe9ec26a02d786451c890636f83f1dc63dc36fdcec4c9526e4
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/qiaodan-part1-30.wav b/tests/test_data/test_audios/10s-mix/qiaodan-part1-30.wav
new file mode 100644
index 0000000000000000000000000000000000000000..0c3f423c009d5fffb5cece918035e679506c784a
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/qiaodan-part1-30.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de45323f5741aac05143b6b995568c8e10e0bad5eadde2df9d67e0770ecc57b8
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/qiaodan-part2-0.wav b/tests/test_data/test_audios/10s-mix/qiaodan-part2-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..0c7b127b648303d6e2ec27400bc81b08af9f26ce
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/qiaodan-part2-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:648770939935801a613bc36917df034811ae70012c72d399582b54ff54d5cae0
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/qiaodan-part2-10.wav b/tests/test_data/test_audios/10s-mix/qiaodan-part2-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..d5635039c60732ed88c627c1d7f716e36d802d99
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/qiaodan-part2-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9b3f117ccbb2bc9bab9355a3174b3fe9b39a6786ad37c0c20d577b643031aff
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/qiaodan-part2-20.wav b/tests/test_data/test_audios/10s-mix/qiaodan-part2-20.wav
new file mode 100644
index 0000000000000000000000000000000000000000..4acaed7f16f2237b3f5cf3aa0bbc087e95719981
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/qiaodan-part2-20.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c918e58777b7d72307089963690a72ee1e67cf733f328fa93a385c204e0e4b02
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/qiaodan-part2-30.wav b/tests/test_data/test_audios/10s-mix/qiaodan-part2-30.wav
new file mode 100644
index 0000000000000000000000000000000000000000..a89e23ece87ef68ea1ff5417695a0ccfce52cb21
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/qiaodan-part2-30.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85891cc64cb79a385d986915dfcfa4fbc3e0ddf12870c7beafcc76c1e012bf78
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/randomforest-part1-0.wav b/tests/test_data/test_audios/10s-mix/randomforest-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..14346661e21333cff352bcb19cb2f06ab99299ad
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/randomforest-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:802125c360c22476bad1aaabb6c5210d21460ed69884f21a3cb318ea00377345
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/randomforest-part1-10.wav b/tests/test_data/test_audios/10s-mix/randomforest-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..4b9cf37e71617db9ca20aa23949ec60a1a290348
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/randomforest-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1605d02574c5a50a91bda5c397079eb6f503311012ed9f15e9b9f90ee7c5f30e
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/randomforest-part1-20.wav b/tests/test_data/test_audios/10s-mix/randomforest-part1-20.wav
new file mode 100644
index 0000000000000000000000000000000000000000..7e2f042b4268601ace23d1ae268027b21ea3db82
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/randomforest-part1-20.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:777b2c46d021c3c10fd0eafce8d6a17bddfc0867119d4537daaff945cf839462
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/randomforest-part1-30.wav b/tests/test_data/test_audios/10s-mix/randomforest-part1-30.wav
new file mode 100644
index 0000000000000000000000000000000000000000..b484e52295f61ea24690a466639a4b218a0fef7f
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/randomforest-part1-30.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9da19f35aab393949a408855ea7970f55ae50f13ceaa0f87e576cb1270cc019
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/zhanghuailong-part1-0.wav b/tests/test_data/test_audios/10s-mix/zhanghuailong-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..c9918aff9685b31fcd9b406147572e91db4fb521
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/zhanghuailong-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6799ca07ab7c8db1ebb97244bc89022b281464dc864de0267b52192db8a3e107
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/zhanghuailong-part1-10.wav b/tests/test_data/test_audios/10s-mix/zhanghuailong-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..dc21055b283f918b494bd7d1c65576d657a85be5
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/zhanghuailong-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8709c003dd26f01d0a140ed3c31e0dd84801984b7a50a36b0239e6b9590538f0
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/zhanghuailong-part1-20.wav b/tests/test_data/test_audios/10s-mix/zhanghuailong-part1-20.wav
new file mode 100644
index 0000000000000000000000000000000000000000..da8eac8f4aa0609297dfd45f616f4a7fce6a44cf
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/zhanghuailong-part1-20.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c15a860a7c6994862244e49cb1fd2e19ac633e1b2c2abddedb10bed480968ed0
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/zhanghuailong-part1-30.wav b/tests/test_data/test_audios/10s-mix/zhanghuailong-part1-30.wav
new file mode 100644
index 0000000000000000000000000000000000000000..6e015698e09d6c9eccc23d7e3599b3df3d3cf19f
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/zhanghuailong-part1-30.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2ecac45c3d44555544ac910a57282c282c6b78950fb96159f736c7d55db7758
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/zhanghuailong-part2-0.wav b/tests/test_data/test_audios/10s-mix/zhanghuailong-part2-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..0606b6922ed94220be6e696dd97a26c82812e13b
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/zhanghuailong-part2-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03cf6c5709a00559d322b190e431a876d791d8d770f2f7116c25520ad43d58f7
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/zhanghuailong-part2-10.wav b/tests/test_data/test_audios/10s-mix/zhanghuailong-part2-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..27a4a8983fbbbe713405b53838dab71190237e35
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/zhanghuailong-part2-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2af66c237c89707ff99a056de8dc2719bf73c5e6e8310ce6371702e669200c2c
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/zhanghuailong-part2-20.wav b/tests/test_data/test_audios/10s-mix/zhanghuailong-part2-20.wav
new file mode 100644
index 0000000000000000000000000000000000000000..d7a0b2fa370f04de15763310a8574f08ff79fa5f
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/zhanghuailong-part2-20.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e768539ebb7a19eacfa3f93670cc65842a5544d7b90719169885eda323a4e4d
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/zhanghuailong-part2-30.wav b/tests/test_data/test_audios/10s-mix/zhanghuailong-part2-30.wav
new file mode 100644
index 0000000000000000000000000000000000000000..27927eabe7a89867a053a6ffe939a57ef8151090
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/zhanghuailong-part2-30.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72978270f9c09934667139af5d04b027674550c7d9b8f0f89e79312f286afd60
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/zhengyaowei-part1-0.wav b/tests/test_data/test_audios/10s-mix/zhengyaowei-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..5b41d76966c89717647f49e6db3d658d49a6008d
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/zhengyaowei-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d05086bbb43e8dde90ed35ab631ad641313c1abd5fc7e735b5dd72c7fa4e486
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/zhengyaowei-part1-10.wav b/tests/test_data/test_audios/10s-mix/zhengyaowei-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..10eb7631ea7b461229e6c79fd42080465aef4cf4
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/zhengyaowei-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd3f87d22212ec4c912119530aa2542bb6a261318935f76637a1a71a441f5e76
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/zhengyaowei-part1-20.wav b/tests/test_data/test_audios/10s-mix/zhengyaowei-part1-20.wav
new file mode 100644
index 0000000000000000000000000000000000000000..1d2be43c5e8bd5647081d035bb2216fdc51ee9f9
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/zhengyaowei-part1-20.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a026310358db67cb2fd7e94c6765db8e54ccbf21685489983c49f3405e2f861
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/zhengyaowei-part1-30.wav b/tests/test_data/test_audios/10s-mix/zhengyaowei-part1-30.wav
new file mode 100644
index 0000000000000000000000000000000000000000..7233467d1bbd2950c8314d2fd0388f3db163c109
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/zhengyaowei-part1-30.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d522330c761a1fdd6f2ee4b8ee57e1ace3d92206f4c0e612861fb78730a1c24
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/zhengyaowei-part2-0.wav b/tests/test_data/test_audios/10s-mix/zhengyaowei-part2-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..9d735c95ef19af803e4a582d702a3029f0e6f7a3
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/zhengyaowei-part2-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67dc5f0fb9c858047877c6df7a814e78119b4062553e9eebe572f849371875b6
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/zhengyaowei-part2-10.wav b/tests/test_data/test_audios/10s-mix/zhengyaowei-part2-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..cb8c1eeaab33b4cc7c219fb0df98f3635eebb077
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/zhengyaowei-part2-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0bdc23285decae69de81fe5cb17e64199e0a500ede6b05ebaa93a223a3394bdc
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/zhengyaowei-part2-20.wav b/tests/test_data/test_audios/10s-mix/zhengyaowei-part2-20.wav
new file mode 100644
index 0000000000000000000000000000000000000000..c99eb7b3aeafba0564f582f998369df10bbbc514
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/zhengyaowei-part2-20.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da5a0d55e6316b696fce61c3749b6fba80ddb76ac934cabf5a5b55f9e698ffbb
+size 320078
diff --git a/tests/test_data/test_audios/10s-mix/zhengyaowei-part2-30.wav b/tests/test_data/test_audios/10s-mix/zhengyaowei-part2-30.wav
new file mode 100644
index 0000000000000000000000000000000000000000..4934c4e48c6ad05a942d60b452049c4386023748
--- /dev/null
+++ b/tests/test_data/test_audios/10s-mix/zhengyaowei-part2-30.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3af78d7025f60f3f3f72e086e6aac59961fb6b19d3e322e75b17e0afd6fdfe7e
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-calculus-part1-0.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-calculus-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..1b6796e045b2779e1a91b90639ef5b4e3bc19124
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-calculus-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7bd7590f5c585bad05b4f71104bd45ca9cd8c0f4091745b52e01612193aa7e6a
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-calculus-part1-10.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-calculus-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..6fccdfc13f7fa10c1a7372a7b6fa9235784424b9
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-calculus-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4546aa10ad12342b587ec4762ef8e061738caba086246098d89a89974ef57a3
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-calculus-part1-20.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-calculus-part1-20.wav
new file mode 100644
index 0000000000000000000000000000000000000000..001cee2329b69fbb53810ab41063fa55c358caad
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-calculus-part1-20.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73d3a949fe4a3e23df41bcd77396b309eaa81e24c995b5fc9f33168475335778
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-calculus-part1-30.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-calculus-part1-30.wav
new file mode 100644
index 0000000000000000000000000000000000000000..f12522aab81f068bf71ec536b89159aa4c38b793
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-calculus-part1-30.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be58b54f49239616fa2bc7c2a1c51544e42850b1575aefe65ffaa69861e31656
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-economics-part1-0.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-economics-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..5b3122270fb9889868e34896172f40e23cea61b3
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-economics-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:acb3e2a28f4d5c75e5937099a2cbfad80a66ed9f1d4a5f3f19cb3ee21e72e2f2
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-economics-part1-10.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-economics-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..57653384217f70252d593c0f9fca68a5e52aca19
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-economics-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c63410f2d817aaeb534d0c898e8b7c5e01a31a621d280536a3a028536263772a
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-economics-part1-20.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-economics-part1-20.wav
new file mode 100644
index 0000000000000000000000000000000000000000..22f7f22baecc18875c7b6bc5920a56f29341e088
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-economics-part1-20.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:620afe0317af35c9bbecfa6fe64d25687808278ac20b2f7534710360074ad4ec
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-economics-part1-30.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-economics-part1-30.wav
new file mode 100644
index 0000000000000000000000000000000000000000..9d236a988ef286cedb4031f185e9235cea3a2ed3
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-economics-part1-30.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd66260b4d092f4beee728eb472cd2499b209b090dc8e3725b4b5bd06ce33240
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-food-part1-0.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-food-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..ffba4142b0badf58c85340e7151a5ff6abbfc590
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-food-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77c114c520b67b2b6f373cf073c2ede13e97b2dc505c3f2a70096cafbab1e1bb
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-food-part1-10.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-food-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..1814e4e1f02a336d7ce949433e47982554c0e41e
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-food-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:898586b7b109d48ebd5ad339058da9fd75d926b0353fe7c58f7302419270c5e6
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-food-part1-20.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-food-part1-20.wav
new file mode 100644
index 0000000000000000000000000000000000000000..e8bef7e92a5f6c233e2e072374ec98ac9a7b0008
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-food-part1-20.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22ddbf31a5f5f179db378e5a182b1f1fed6125c36484bb5c8067fd73da1b074d
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-food-part1-30.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-food-part1-30.wav
new file mode 100644
index 0000000000000000000000000000000000000000..b4ddfe297283e6ec9f9a4052552137df60bb84c1
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-food-part1-30.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac38cfb57d2187cc677903e58ff1dee6b4685dd924c3b91a5254291a876e392a
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-healthandweather-part1-0.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-healthandweather-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..b584e782f670d5573f974faefff3ea8db247bc8a
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-healthandweather-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd43697595ac1429c60b37be49bc83a938872cca7e2902ce239a210d07a466ad
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-healthandweather-part1-10.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-healthandweather-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..3032d69e1e7db463b8dea472546d7e21e91469af
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-healthandweather-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8099068117172d7492e03f9428fc97638ffbc0bf16591256911b8db313c6fed1
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-healthandweather-part1-20.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-healthandweather-part1-20.wav
new file mode 100644
index 0000000000000000000000000000000000000000..0e219a3dabb206da0d595b23797f9143ce3303d7
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-healthandweather-part1-20.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:311e20176b6eab2d4f8277ff1b5fefe67ecb5d5401f8bcc6117f9605e3e76863
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-healthandweather-part1-30.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-healthandweather-part1-30.wav
new file mode 100644
index 0000000000000000000000000000000000000000..5c900dcb3920ee54d44856e1de6b73cbab98d786
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-healthandweather-part1-30.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3dd5c869a7b24e1ff00a7e728fa356f8a497cd4632b6922580d5b284ca71a8bb
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-laws-part1-0.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-laws-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..20df925f85a14d0b784d7b99e33d4167e665cf4c
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-laws-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:78c511ab1f4af1723faeec23922441fb0812f666c6b268669579532534f78ed5
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-laws-part1-10.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-laws-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..8c06cef7c00863fa5c4d9189efbbe076d8d7996c
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-laws-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2cc21c84d66486701fbba6ced5c90a45e1f1216feb9ecb101530dcf91bdc7d7b
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-laws-part1-20.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-laws-part1-20.wav
new file mode 100644
index 0000000000000000000000000000000000000000..d8b1b442909ef0dc1b8bdaa284ed186a8d49d5d1
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-laws-part1-20.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5cb3deac03549f0464cc7425eb55b7a2851599e1188f1bf62b3c540718902c6f
+size 320078
diff --git a/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-laws-part1-30.wav b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-laws-part1-30.wav
new file mode 100644
index 0000000000000000000000000000000000000000..e66b864e8e752dff9367e4efc2b4e979c771b843
--- /dev/null
+++ b/tests/test_data/test_audios/10s-zh-ac1-16k/Chinese-laws-part1-30.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4881819f902004ada1e49f2b39f2aaa684c65b1a8b34be09b842b141a2fe0d0
+size 320078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-chaos-part2-0.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-chaos-part2-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..1b465e178461c1c47f58d97ba8b65b6f14ee5a48
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-chaos-part2-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f73ba6127284a462aca3a04f66efbb33fc4a12717428b42bdf505b2536d3b3ef
+size 160078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-chaos-part2-10.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-chaos-part2-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..8e351493525a69ef7544a061a106b99a5173848b
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-chaos-part2-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c633c887127f299ed022d69fc4be2802dd0b5c99a0acd3132df88d05ea0243d
+size 160078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-chaos-part2-15.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-chaos-part2-15.wav
new file mode 100644
index 0000000000000000000000000000000000000000..8db60e5f3627b83e15e46d2a3917efbf6dde28ae
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-chaos-part2-15.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bce2151c86797f24af02a8c84feb6aab21ba7127430ee06899cc4736eb3b3051
+size 160078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-chaos-part2-5.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-chaos-part2-5.wav
new file mode 100644
index 0000000000000000000000000000000000000000..c89ce75be6338900697c5c6590737dbf06d62f83
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-chaos-part2-5.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d5ea44be5e69f713c0cd996d62f3c1a843177eb2418497ac8ce3d7d4bd02113
+size 160078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-computer_sicence-part1-0.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-computer_sicence-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..d7d11f4d3388532a68cfeb1ca785b0c33a37d7d6
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-computer_sicence-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ece9a71924cce20d2784277840c846edb7a837a984dc32d02f590cd7706d01d5
+size 160078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-computer_sicence-part1-10.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-computer_sicence-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..78bc5bfa8f3eaab09cd71ed3cf614e496c1d0bdf
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-computer_sicence-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:629093d391aeec3d90236cbf1d75875916f659b86701a1322cf01c50eef8169c
+size 160078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-computer_sicence-part1-15.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-computer_sicence-part1-15.wav
new file mode 100644
index 0000000000000000000000000000000000000000..a87082839971e20ec080243914354cc3d06e7f95
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-computer_sicence-part1-15.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9bc63d5a123f6ff4c0308e6290be1e330b0e12a3d219103873474bcdfd9e8aa
+size 160078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-computer_sicence-part1-5.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-computer_sicence-part1-5.wav
new file mode 100644
index 0000000000000000000000000000000000000000..3caf109f9a21f3e04c56c5f63b18905f86b67fb3
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-computer_sicence-part1-5.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efcd22480f332430180528760fded66fdae298588e6befd088cf7073fda93bc7
+size 160078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-generative_ai-part1-0.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-generative_ai-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..425550302615711851ecb20d6cc73c9350035dfd
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-generative_ai-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05f056520112a476437f558e8a3bdbfaab7a87755a4444c8f047ce5aced464ea
+size 160078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-generative_ai-part1-10.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-generative_ai-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..fda281b839639d1536c3ee482f69327475966f13
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-generative_ai-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a3f416a5aaded57d452720061c10a0c98f3e4f56a118477c82bae2ef17eebdd
+size 160078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-generative_ai-part1-15.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-generative_ai-part1-15.wav
new file mode 100644
index 0000000000000000000000000000000000000000..99b940f4a1f1d747e52f6990a87256c47bab2d67
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-generative_ai-part1-15.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db7336af608276c1cb7d6adcc75f2c4685facb945bac573c11e4ff876b58bc43
+size 160078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-generative_ai-part1-5.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-generative_ai-part1-5.wav
new file mode 100644
index 0000000000000000000000000000000000000000..bb666b0fe2e3bcafe27d56fee65e04438a19b4e5
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-generative_ai-part1-5.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81b5b891ce9a6bbe409bb37836599e361c836433196f11deaf7fcc337509b3c5
+size 160078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-internet-part20-0.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-internet-part20-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..87d8c35490025309b95dc7b9e8bea3d14c02a5a7
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-internet-part20-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51ac2b9a997a94173e966ecb4868e9abdd6ebd857f13791a4ce092959c4abb35
+size 160078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-internet-part20-10.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-internet-part20-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..24ec0c0963670a63898067de4670848ac5bf63f5
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-internet-part20-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7ce8be01696bfd51c516ecea86b69b4f6593d947020013ed86a081266b184b8
+size 160078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-internet-part20-15.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-internet-part20-15.wav
new file mode 100644
index 0000000000000000000000000000000000000000..6fac24b129bb6502ae40d222ea28a15e24b6c7df
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-internet-part20-15.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3bdd370f994e73949829bdd8baa8bf65bceca3f86229e94883ca0058462fa07d
+size 160078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-internet-part20-5.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-internet-part20-5.wav
new file mode 100644
index 0000000000000000000000000000000000000000..9a01a6fa449c912a801281fa7c1ac4e8ce47e72c
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-internet-part20-5.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2187751ebf84495137f21c00c406646fc076b44566010db24961506b93e7de4e
+size 160078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-legalsystem-part1-0.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-legalsystem-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..3c074ff59eaeec44d81edd9df74c3e5efafcd9be
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-legalsystem-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a473ed09c983c68e0d1e40a292d3c01440a266c171d31d42e927b4a3c0433650
+size 160078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-legalsystem-part1-10.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-legalsystem-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..8d17fc019251c5e0b60904a50934e7ffaa9aa326
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-legalsystem-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:276d86776b9a66e614178e9b9c16eba9a72ba807070474b28955795145a16775
+size 160078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-legalsystem-part1-15.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-legalsystem-part1-15.wav
new file mode 100644
index 0000000000000000000000000000000000000000..418ab07b9a031debc5c20f71dbe5723b7345d34b
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-legalsystem-part1-15.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2316e42b1bc04511f683288a7063fd8da9d84681791add45879352a25d1976c
+size 160078
diff --git a/tests/test_data/test_audios/5s-en-ac1-16k/English-legalsystem-part1-5.wav b/tests/test_data/test_audios/5s-en-ac1-16k/English-legalsystem-part1-5.wav
new file mode 100644
index 0000000000000000000000000000000000000000..f42f6809c2cc36de2b454a2ff9d249ba4297cda7
--- /dev/null
+++ b/tests/test_data/test_audios/5s-en-ac1-16k/English-legalsystem-part1-5.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7cbadc61dbe77b0278e23d8dc12e4aec4d864867bcc38fbf0d92a44d5215f2e7
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/qiaodan-part1-0.wav b/tests/test_data/test_audios/5s-mix/qiaodan-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..f998f2516bb23327ecfa85f9661e8ac229741df5
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/qiaodan-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84b711d9298430af7ffbe31a8ed743f0b91766c14078aa4b2fd712badcb3e043
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/qiaodan-part1-10.wav b/tests/test_data/test_audios/5s-mix/qiaodan-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..ba5fca98872ff8d337da568e8fdb4fcb3a3734b5
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/qiaodan-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:818dcd7556a316a11413f9c3188f6128e24496ca9199ec11dc0cc510d96a5474
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/qiaodan-part1-15.wav b/tests/test_data/test_audios/5s-mix/qiaodan-part1-15.wav
new file mode 100644
index 0000000000000000000000000000000000000000..526571c414836a79163fecdd4a6bab5ab6f599aa
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/qiaodan-part1-15.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1702c178d285ded2fe4b2222e79cd9e4a61b8bdae1b67ece0cce67525269e5de
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/qiaodan-part1-5.wav b/tests/test_data/test_audios/5s-mix/qiaodan-part1-5.wav
new file mode 100644
index 0000000000000000000000000000000000000000..4671f896d1d9a042258b3617d7ff1a217f8f9f97
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/qiaodan-part1-5.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79f5616d13a29a1e61385b06173a7f7fe1fa2eaf01ee5762b6d057e707115b76
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/qiaodan-part2-0.wav b/tests/test_data/test_audios/5s-mix/qiaodan-part2-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..aa66fdfc3fd69f17cae47f313b1bff3242954ee8
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/qiaodan-part2-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:358e3657d329d80239c8877609ad18e811fb1edf54fe99583d7089dd4a4d94a1
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/qiaodan-part2-10.wav b/tests/test_data/test_audios/5s-mix/qiaodan-part2-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..77151f45cd095b6a7c520ddab1c9f2a419ef652c
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/qiaodan-part2-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e161854b6880b4601b7de829000d54b6f9ee0ce0d9dd5a2b489d1f89a4ccc5c2
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/qiaodan-part2-15.wav b/tests/test_data/test_audios/5s-mix/qiaodan-part2-15.wav
new file mode 100644
index 0000000000000000000000000000000000000000..da102286c42a4efaa4cb1fea1c652fef4b0ccc1c
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/qiaodan-part2-15.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9466ff77df8a6a2b96279d16b4b4998ee96b87f2987dbb5704b3f6b87cedfad4
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/qiaodan-part2-5.wav b/tests/test_data/test_audios/5s-mix/qiaodan-part2-5.wav
new file mode 100644
index 0000000000000000000000000000000000000000..31897bc6e6e3341198a1ea02f2485943b7731115
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/qiaodan-part2-5.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d9277158b49b75a26c47b56dec5b1dcdd35dbf69b7c215b36fc69c71274de2f
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/randomforest-part1-0.wav b/tests/test_data/test_audios/5s-mix/randomforest-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..ee8f4d31ddf32909c7ad159c72fafa45f60741e6
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/randomforest-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6545655580d8e5d114382e32ea2819807a3a30f2b9cef7d13b8b52b6fbef692
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/randomforest-part1-10.wav b/tests/test_data/test_audios/5s-mix/randomforest-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..1117742efccf31e7e789ebb622753eea71b7d576
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/randomforest-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c094d7b81e7ddc484bb338c66bfa08e09cc8912965013a69ec8eabd7ab1e6e7
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/randomforest-part1-15.wav b/tests/test_data/test_audios/5s-mix/randomforest-part1-15.wav
new file mode 100644
index 0000000000000000000000000000000000000000..f8845aacd755cd46637ab3e8ff87f16b086a2c00
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/randomforest-part1-15.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bddf229d4cafe1c6e4dfc774586b80a4c48f211712b9a77768ac2b11f5c88c87
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/randomforest-part1-5.wav b/tests/test_data/test_audios/5s-mix/randomforest-part1-5.wav
new file mode 100644
index 0000000000000000000000000000000000000000..031da5ac9b0902a299e141b7ef9400540415402a
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/randomforest-part1-5.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c35f872b00498ff591d5ccb46aef9832595ea58204735916a419da3acefcb0c
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/zhanghuailong-part1-0.wav b/tests/test_data/test_audios/5s-mix/zhanghuailong-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..eacf0cc11d2a5b87b4b5fe8c182168c32ad495db
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/zhanghuailong-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e3e394fb6a3af16ba8b3546fa116b6a17cc9844887ac8cb570d28a370fd09a2
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/zhanghuailong-part1-10.wav b/tests/test_data/test_audios/5s-mix/zhanghuailong-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..fcca68f096e2e31f1800807f548a1bf2381b5749
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/zhanghuailong-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b3babb795dd4017c9e7e0e3f2d8017def0c2e5b57794064f6aa740fb32193004
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/zhanghuailong-part1-15.wav b/tests/test_data/test_audios/5s-mix/zhanghuailong-part1-15.wav
new file mode 100644
index 0000000000000000000000000000000000000000..4d8ab8f7179fcd6e4b20918c47c6ff00dd8c931d
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/zhanghuailong-part1-15.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80058c7e056fe0de7291f486dd7a3945d3434437b955505de54432083e9b146c
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/zhanghuailong-part1-5.wav b/tests/test_data/test_audios/5s-mix/zhanghuailong-part1-5.wav
new file mode 100644
index 0000000000000000000000000000000000000000..4941e37afd743dd5d8e93ade171938d795e4d7da
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/zhanghuailong-part1-5.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f1a7cb1b4ae9e140c2c9f67cc3db4e1b625b5717c47ff8d6a10d4b5dd824c15
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/zhanghuailong-part2-0.wav b/tests/test_data/test_audios/5s-mix/zhanghuailong-part2-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..4e6896c16d01b8daa30e57bbf20aff0466b9c083
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/zhanghuailong-part2-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ddab9cd2c80af1500edfcace6c9b5500c3bc8c71002e42d6a4be11cf86da191
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/zhanghuailong-part2-10.wav b/tests/test_data/test_audios/5s-mix/zhanghuailong-part2-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..0960e937471408b1ab0bfb785f9f1f317ef9ee19
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/zhanghuailong-part2-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8263399481e72884d1d3f678d9ecc4be0256fa5984d2ac8dbb2c40504350e588
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/zhanghuailong-part2-15.wav b/tests/test_data/test_audios/5s-mix/zhanghuailong-part2-15.wav
new file mode 100644
index 0000000000000000000000000000000000000000..0e78bad8e9fce15f867fd40e591eb31544cc80c3
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/zhanghuailong-part2-15.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2159ea09febaa04a1dbe393ff84020cf65f1425409ab4e4f21109c1a9a49fef5
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/zhanghuailong-part2-5.wav b/tests/test_data/test_audios/5s-mix/zhanghuailong-part2-5.wav
new file mode 100644
index 0000000000000000000000000000000000000000..0fec4f3e7912935079bcb4e3fb1591113fc2c750
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/zhanghuailong-part2-5.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd055c04537d3c7c6bc1a0bbdd9459395409290be489a067b495c54942a29b69
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/zhengyaowei-part1-0.wav b/tests/test_data/test_audios/5s-mix/zhengyaowei-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..292f7327be4c6a6438b9e8b21985a75633d51324
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/zhengyaowei-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c6e4b7a2618c706b1f95098777fdb907bcbb23ebf79c384c841359791295f2d
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/zhengyaowei-part1-10.wav b/tests/test_data/test_audios/5s-mix/zhengyaowei-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..6129662829eefc09c338bd93c4a7792d8a861446
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/zhengyaowei-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:422d62c288cb76db85f5d436cbb9ba646ece2ef13bc7920adc8952d675cf990d
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/zhengyaowei-part1-15.wav b/tests/test_data/test_audios/5s-mix/zhengyaowei-part1-15.wav
new file mode 100644
index 0000000000000000000000000000000000000000..f74431a13182c0792e9e48e490c3a751a28d7001
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/zhengyaowei-part1-15.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:297896a49cbbce48bad94bfb3c0eb0253f7932dee67126e95ce4a8f8dbed706b
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/zhengyaowei-part1-5.wav b/tests/test_data/test_audios/5s-mix/zhengyaowei-part1-5.wav
new file mode 100644
index 0000000000000000000000000000000000000000..9373e36cc04788d6c56931967125daa0ca5d789b
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/zhengyaowei-part1-5.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:12116103fac0ed68321f821d3b398aa825663a00ea98d059f504f9c3d5eee897
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/zhengyaowei-part2-0.wav b/tests/test_data/test_audios/5s-mix/zhengyaowei-part2-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..248defa385ff70d33ee2954734c52bb944ea85f9
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/zhengyaowei-part2-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6ea0669e3605bff08ca424d8f2df610419305f1f77dc90daaebbf96725d30c7
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/zhengyaowei-part2-10.wav b/tests/test_data/test_audios/5s-mix/zhengyaowei-part2-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..4a3fa9adb18a7a5ef64450c5003802c98a656c9c
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/zhengyaowei-part2-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:634a8a319c87e877851dfc49ab766136b2d565b229c7f0d41f52985fc85c8306
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/zhengyaowei-part2-15.wav b/tests/test_data/test_audios/5s-mix/zhengyaowei-part2-15.wav
new file mode 100644
index 0000000000000000000000000000000000000000..b09012763844d2ef6f3dfe73893fa96653d69eb5
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/zhengyaowei-part2-15.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:438b4ac06d9244a5c250f1ea820a9f9aa184d27d8b39932e8564adc31d2203c7
+size 160078
diff --git a/tests/test_data/test_audios/5s-mix/zhengyaowei-part2-5.wav b/tests/test_data/test_audios/5s-mix/zhengyaowei-part2-5.wav
new file mode 100644
index 0000000000000000000000000000000000000000..1c6017362112b87bec829e99ccfb0157e3464020
--- /dev/null
+++ b/tests/test_data/test_audios/5s-mix/zhengyaowei-part2-5.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a82d43ee2981f73c63c17fb3cbc5240ea6468722bfe9ec735032a01482a111a
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-calculus-part1-0.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-calculus-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..15a36de9ce2dc78038e573ae7664d2ba1f0fd9f9
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-calculus-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d07f92e3134255244105fa03990d856c02bf0f06d7ffb1a77f3e058f245fad1b
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-calculus-part1-10.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-calculus-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..fa57c8f83d6bf1d5e2b22479b594066d9a220018
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-calculus-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0a3b25ec29e71494ff3f4a686599b0a36260e4b323ddde07818ab570155242a
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-calculus-part1-15.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-calculus-part1-15.wav
new file mode 100644
index 0000000000000000000000000000000000000000..e4524c9502072714b0d254f6b33a5e12549b34ad
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-calculus-part1-15.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c2929b5c292b6c8901587035faa5a32cd7383baeeac1cf34b2079d07ceb90de
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-calculus-part1-5.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-calculus-part1-5.wav
new file mode 100644
index 0000000000000000000000000000000000000000..ecd4741309d23724733850291b329e85f8925513
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-calculus-part1-5.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7ca924891f33d3873c252e80914024499673571b57ccb10dc2422d4cd2c5e11
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-economics-part1-0.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-economics-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..fb7d8b31a28535a5d6c3281578b07364a30fbbda
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-economics-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a246de41f5e720f4e020fbb38cedf0f9b9a6f2c9fa5a8d2f40d358bfe791b89f
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-economics-part1-10.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-economics-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..2eff33b18f58c5546df7070910539b04c35bdb51
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-economics-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9157e84e85cdb4cc4fe884643074a348b63888bf67cc262396ab90ce5e847e20
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-economics-part1-15.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-economics-part1-15.wav
new file mode 100644
index 0000000000000000000000000000000000000000..a593baba6478377d09eb6fa049add907268ac3cf
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-economics-part1-15.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13b3ac6047d7cb52926a0826f9a06f8d02142d2118235e360faa3b52ea5d428a
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-economics-part1-5.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-economics-part1-5.wav
new file mode 100644
index 0000000000000000000000000000000000000000..a7157c5deaca9a7f7e5980ef820d1dc943849b8e
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-economics-part1-5.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f171bd647c28b342a3bf39d93d30e4a63634221e03dc107ee7d6f9ccf76d5428
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-food-part1-0.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-food-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..ad78e67fc3a34105aa506d93a01b249b6016e36c
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-food-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9cf9a1a252ffa7588f08cb53fd978c1f5043fd0341eb21777f9e5239f3e54ed0
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-food-part1-10.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-food-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..11b82736732badd5ca87fc987c861c46570efcff
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-food-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fbdf50e1d5e2a963a631ac2875cef4689d6b73bf9a1c021860870f57c11732e9
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-food-part1-15.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-food-part1-15.wav
new file mode 100644
index 0000000000000000000000000000000000000000..8e55ff1191447c9490ddfc8908d71460eb0758aa
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-food-part1-15.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b67138119fdbc82afaf6c4538c80ebe6c0d5a68ff780a386147bfdf735b2ff7
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-food-part1-5.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-food-part1-5.wav
new file mode 100644
index 0000000000000000000000000000000000000000..24509760aacb1241ce012cb999ac322e4319d7d8
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-food-part1-5.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46084c2d8a905cd1e7f6e8b0ca920eaf0eefdb052f5a60e23d1abc4507462627
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-healthandweather-part1-0.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-healthandweather-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..230ff3ef102a0f9ed1804942b2e0134fbac07c8f
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-healthandweather-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6bffff46e8bb26c59d48a66b7d887bf705f57fe80975917b4658110999b1d610
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-healthandweather-part1-10.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-healthandweather-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..22bc6b35d23648a6f0715c0e8e9d4db062ef7d28
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-healthandweather-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ecfbc3c21bdd9aad2c26acd546a75c5d7fea735e11d0549bb21179a6f70de34
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-healthandweather-part1-15.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-healthandweather-part1-15.wav
new file mode 100644
index 0000000000000000000000000000000000000000..13e7186d1cf66b0ea2374f22adef0a2852c5cd91
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-healthandweather-part1-15.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:012bba9bb45369d2a496f424ead452b315fa09dcbd1a00eb84e2d633bbb28d58
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-healthandweather-part1-5.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-healthandweather-part1-5.wav
new file mode 100644
index 0000000000000000000000000000000000000000..20bf4ca88991c10aae80815bae50868d61535cd3
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-healthandweather-part1-5.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29e62696ff361d2a98883cf84cf99c2254d94703c9c7b116405f0666650c6c4a
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-laws-part1-0.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-laws-part1-0.wav
new file mode 100644
index 0000000000000000000000000000000000000000..5096830c9bf0be66f0eb44237b17273abd04fd26
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-laws-part1-0.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d7e6a4741f25e0c52b87abd1530f86c454fbc9f116ace0fe149954809259639
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-laws-part1-10.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-laws-part1-10.wav
new file mode 100644
index 0000000000000000000000000000000000000000..089820da45ce066bdafb158174e9d07eca4e67ea
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-laws-part1-10.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91386e2e469fc3d3d189fb0685e55dda7acb2f3291a0e80c1ec2a2b6759e0d46
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-laws-part1-15.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-laws-part1-15.wav
new file mode 100644
index 0000000000000000000000000000000000000000..bede48dd91d582aaf5319114ace79c356e2656c1
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-laws-part1-15.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53f5b8515932117a9df5395930d7b881915cf10062b620de04fd6f58fbf8b2e7
+size 160078
diff --git a/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-laws-part1-5.wav b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-laws-part1-5.wav
new file mode 100644
index 0000000000000000000000000000000000000000..1592080886834ae0a575a9607a931738afc12cce
--- /dev/null
+++ b/tests/test_data/test_audios/5s-zh-ac1-16k/Chinese-laws-part1-5.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ffd11c53d0d7dda33e4f9ef5f7dd16b58034b78a9d85946d14ef14656d8d8c61
+size 160078
diff --git a/tests/test_data/test_audios/qiaodan-part1.mp3 b/tests/test_data/test_audios/qiaodan-part1.mp3
new file mode 100644
index 0000000000000000000000000000000000000000..3f394aee8e5734be4814e3aa1e36603c6584aee4
--- /dev/null
+++ b/tests/test_data/test_audios/qiaodan-part1.mp3
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3009a99097646c8c84aa784ef094f2267520719699c4e78072a1dcde10820b31
+size 1072829
diff --git a/tests/test_data/test_audios/qiaodan-part2.mp3 b/tests/test_data/test_audios/qiaodan-part2.mp3
new file mode 100644
index 0000000000000000000000000000000000000000..b4024686e954028e2165ab30472d0b769181daa0
--- /dev/null
+++ b/tests/test_data/test_audios/qiaodan-part2.mp3
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7755f1bbfd1e423969271d5c2d32bd2f4e9ec4fb0310dd7293fe654921e373a7
+size 992999
diff --git a/tests/test_data/test_audios/randomforest-part1.wav b/tests/test_data/test_audios/randomforest-part1.wav
new file mode 100644
index 0000000000000000000000000000000000000000..0d2c6e5eb22dd65bd471243084cde02bff1f8b15
--- /dev/null
+++ b/tests/test_data/test_audios/randomforest-part1.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2dc09f7c3e81f600c98a5b0143ae201dab8672a46d4c00d9ac0f07ff6f3c69b
+size 1920078
diff --git a/tests/test_data/test_audios/zhanghuailong-part1.mp3 b/tests/test_data/test_audios/zhanghuailong-part1.mp3
new file mode 100644
index 0000000000000000000000000000000000000000..e90f3ebdb6d64fbec335ff36a3355d2bb0550d54
--- /dev/null
+++ b/tests/test_data/test_audios/zhanghuailong-part1.mp3
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d1efce201560e8dea30d843e233f12936fbfe000e309558312b5c90752100cb3
+size 929051
diff --git a/tests/test_data/test_audios/zhanghuailong-part2.mp3 b/tests/test_data/test_audios/zhanghuailong-part2.mp3
new file mode 100644
index 0000000000000000000000000000000000000000..36cbf9565a9d2bedda86569bbc925f87dadc1e17
--- /dev/null
+++ b/tests/test_data/test_audios/zhanghuailong-part2.mp3
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af67d760b7d350a9f153f93219c3e1d80942cedcb9b4b44f8f6ae5dfa6f4a1a7
+size 1072829
diff --git a/tests/test_data/test_audios/zhengyaowei-part1.mp3 b/tests/test_data/test_audios/zhengyaowei-part1.mp3
new file mode 100644
index 0000000000000000000000000000000000000000..dbf9d575fd1f494a916252355aef0c1b3ebf126b
--- /dev/null
+++ b/tests/test_data/test_audios/zhengyaowei-part1.mp3
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca8c72864b62b5d867e4eeea6b8708b0b2e7f4bcc2eb240b7f8caf4d86cd9686
+size 1184842
diff --git a/tests/test_data/test_audios/zhengyaowei-part2.mp3 b/tests/test_data/test_audios/zhengyaowei-part2.mp3
new file mode 100644
index 0000000000000000000000000000000000000000..ab222c1728d3408be699e22719b47f6e52897080
--- /dev/null
+++ b/tests/test_data/test_audios/zhengyaowei-part2.mp3
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9060e0b4ce9cb362e4837cad0e050ae0391f97e237a2f9234029a51cf057c071
+size 2065064