Found a bug while loading model "tiny.en" and "small.en" #1421
-
I think I found a bug or error while loading these 2 model: "tiny.en" and "small.en" Here is the code: import whisper
import traceback
model = "tiny"
language = "en"
audio = "./Recording.wav"
if model != "large" and language == "en":
model = model + ".en"
try:
loaded_model = whisper.load_model(model)
except Exception as e:
print(f"Failed to load model: {model}. Exception: {str(e)}")
traceback.print_exc()
text = loaded_model.transcribe(audio=audio, language=language)
print(text["text"]) Error I have tested base, medium and large model works fine |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments 1 reply
-
cannot reproduce the error are you running multiple python sessions ? the error |
Beta Was this translation helpful? Give feedback.
-
The error fixed: D:\Coding\envs\STT\lib\site-packages\whisper_init_.py:63: UserWarning: C:\Users\rudyr.cache\whisper\small.en.pt exists, but the SHA256 checksum does not match; re-downloading the file |
Beta Was this translation helpful? Give feedback.
The error fixed:
D:\Coding\envs\STT\lib\site-packages\whisper_init_.py:63: UserWarning: C:\Users\rudyr.cache\whisper\small.en.pt exists, but the SHA256 checksum does not match; re-downloading the file