Update app.py

app.py

@@ -19,8 +19,9 @@ asr = pipeline("automatic-speech-recognition", model="jonatasgrosman/wav2vec2-la
 
 
 
-model = AutoModelForSeq2SeqLM.from_pretrained('hackathon-pln-es/t5-small-spanish-nahuatl')
-tokenizer = AutoTokenizer.from_pretrained('hackathon-pln-es/t5-small-spanish-nahuatl')
+model_name = 'hackathon-pln-es/t5-small-finetuned-spanish-to-quechua'
+model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
+tokenizer = AutoTokenizer.from_pretrained(model_name)
 
 new_line = '\n'
 
@@ -28,9 +29,9 @@ def predict_and_ctc_lm_decode(input_file):
     speech = load_and_fix_data(input_file, sampling_rate)
     transcribed_text = asr(speech, chunk_length_s=5, stride_length_s=1)
     transcribed_text = transcribed_text["text"]
-    input_ids = tokenizer('translate Spanish to Nahuatl: ' + transcribed_text, return_tensors='pt').input_ids
-    outputs = model.generate(input_ids, max_length=512)
-    outputs = tokenizer.batch_decode(outputs, skip_special_tokens=True)[0]
+    input = tokenizer(transcribed_text, return_tensors="pt")
+    output = model.generate(input["input_ids"], max_length=40, num_beams=4, early_stopping=True)
+    output = tokenizer.decode(output[0], skip_special_tokens=True)
     return f"Spanish Audio Transcription: {transcribed_text} {new_line} Nahuatl Translation :{outputs}"