Update lid.py

lid.py

@@ -2,77 +2,78 @@ from transformers import Wav2Vec2ForSequenceClassification, AutoFeatureExtractor
 import torch
 import librosa
 import numpy as np
+import os
 
-model_id = "facebook/mms-lid-1024"
-
-processor = AutoFeatureExtractor.from_pretrained(model_id)
-model = Wav2Vec2ForSequenceClassification.from_pretrained(model_id)
-
+# Load Facebook MMS Language Identification Model
+MODEL_ID = "facebook/mms-lid-1024"
+processor = AutoFeatureExtractor.from_pretrained(MODEL_ID)
+model = Wav2Vec2ForSequenceClassification.from_pretrained(MODEL_ID)
 
+# Constants
 LID_SAMPLING_RATE = 16_000
-LID_TOPK = 10
-LID_THRESHOLD = 0.33
-
+LID_THRESHOLD = 0.33  # Confidence threshold
 LID_LANGUAGES = {}
-with open(f"data/lid/all_langs.tsv") as f:
+
+# Load Language Labels
+LANG_FILE = "data/lid/all_langs.tsv"
+if not os.path.exists(LANG_FILE):
+    raise FileNotFoundError(f"Language file '{LANG_FILE}' not found!")
+
+with open(LANG_FILE, encoding="utf-8") as f:
     for line in f:
-        iso, name = line.split(" ", 1)
+        iso, name = line.strip().split(" ", 1)
         LID_LANGUAGES[iso] = name
 
-
-def identify(audio_data = None):
+# Identify Audio Language
+def identify(audio_data=None):
     if not audio_data:
         return "<<ERROR: Empty Audio Input>>"
-        
+    
+    # Microphone Input
     if isinstance(audio_data, tuple):
-        # microphone
         sr, audio_samples = audio_data
         audio_samples = (audio_samples / 32768.0).astype(np.float32)
         if sr != LID_SAMPLING_RATE:
-            audio_samples = librosa.resample(
-                audio_samples, orig_sr=sr, target_sr=LID_SAMPLING_RATE
-            )
-    else:
-        # file upload
-        isinstance(audio_data, str)
-        audio_samples = librosa.load(audio_data, sr=LID_SAMPLING_RATE, mono=True)[0]
+            audio_samples = librosa.resample(audio_samples, orig_sr=sr, target_sr=LID_SAMPLING_RATE)
 
-    inputs = processor(
-        audio_samples, sampling_rate=LID_SAMPLING_RATE, return_tensors="pt"
-    )
+    # File Upload
+    elif isinstance(audio_data, str):
+        if not os.path.exists(audio_data):
+            return f"<<ERROR: File '{audio_data}' not found!>>"
+        audio_samples, _ = librosa.load(audio_data, sr=LID_SAMPLING_RATE, mono=True)
 
-    # set device
-    if torch.cuda.is_available():
-        device = torch.device("cuda")
-    elif (
-        hasattr(torch.backends, "mps")
-        and torch.backends.mps.is_available()
-        and torch.backends.mps.is_built()
-    ):
-        device = torch.device("mps")
     else:
-        device = torch.device("cpu")
+        return "<<ERROR: Invalid Audio Input>>"
+
+    # Process Input
+    inputs = processor(audio_samples, sampling_rate=LID_SAMPLING_RATE, return_tensors="pt")
 
+    # Select Device
+    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
     model.to(device)
     inputs = inputs.to(device)
 
+    # Predict Language
     with torch.no_grad():
         logit = model(**inputs).logits
 
+    # Compute Probabilities
     logit_lsm = torch.log_softmax(logit.squeeze(), dim=-1)
     scores, indices = torch.topk(logit_lsm, 5, dim=-1)
-    scores, indices = torch.exp(scores).to("cpu").tolist(), indices.to("cpu").tolist()
+    scores, indices = torch.exp(scores).cpu().tolist(), indices.cpu().tolist()
+
+    # Map to Language Labels
     iso2score = {model.config.id2label[int(i)]: s for s, i in zip(scores, indices)}
+
+    # Confidence Check
     if max(iso2score.values()) < LID_THRESHOLD:
-        return "Low confidence in the language identification predictions. Output is not shown!"
-    return {LID_LANGUAGES[iso]: score for iso, score in iso2score.items()}
+        return "Low confidence in language detection. No output shown."
 
+    return {LID_LANGUAGES.get(iso, iso): score for iso, score in iso2score.items()}
 
+# Example Usage
 LID_EXAMPLES = [
     ["upload/english.mp3"],
     ["upload/tamil.mp3"],
     ["upload/burmese.mp3"],
 ]
-    demo.launch()
-).launch()
-    demo.launch()