STT_Server/tests/modelsuse.py

37 lines
1.3 KiB
Python

from funasr import AutoModel
from typing import List, Dict, Any
from src.pydantic_models import VADResponse
import time
def vad_model_use_online(file_path: str) -> List[Dict[str, Any]]:
chunk_size = 100 # ms
model = AutoModel(model="fsmn-vad", model_revision="v2.0.4", disable_update=True)
vad_result = VADResponse()
vad_result.time_chunk_index_callback = lambda index: print(f"回调: {index}")
items = []
import soundfile
speech, sample_rate = soundfile.read(file_path)
chunk_stride = int(chunk_size * sample_rate / 1000)
cache = {}
total_chunk_num = int(len((speech)-1)/chunk_stride+1)
for i in range(total_chunk_num):
time.sleep(0.1)
speech_chunk = speech[i*chunk_stride:(i+1)*chunk_stride]
is_final = i == total_chunk_num - 1
res = model.generate(input=speech_chunk, cache=cache, is_final=is_final, chunk_size=chunk_size)
if len(res[0]["value"]):
vad_result += VADResponse.from_raw(res)
for item in res[0]["value"]:
items.append(item)
vad_result.process_time_chunk()
# for item in items:
# print(item)
return vad_result
if __name__ == "__main__":
vad_result = vad_model_use_online("tests/vad_example.wav")
# print(vad_result)