save

app.py

@@ -7,7 +7,7 @@ import requests
 import time
 from transformers import pipeline
 
-model = whisper.load_model("small")
+model = whisper.load_model("small.en")
 pipe = pipeline(
     "automatic-speech-recognition",
     model="openai/whisper-small",
@@ -41,7 +41,7 @@ def runApi():
 
     audio = "audio.mp3"
 
-    audio = whisper.load_audio(audio)
+    # audio = whisper.load_audio(audio)
     # audio = whisper.pad_or_trim(audio)
     
     # mel = whisper.log_mel_spectrogram(audio).to(model.device)
@@ -51,14 +51,15 @@ def runApi():
     # options = whisper.DecodingOptions(fp16 = False)
     # result = whisper.decode(model, mel, options)
 
-    result = pipe(audio.copy())["text"]
+    # result = pipe(audio.copy())["text"]
+    result = model.transcribe(audio)
 
     end_time = time.time()
     total_time = end_time - start_time
 
     return jsonify({
         "audio_url": audio_url,
-        "result": result,
+        "result": result['text'],
         "exec_time_sec": total_time
     })