added english transcription tabs

app.py

@@ -8,6 +8,7 @@ import time
 
 p_ta = pipeline("automatic-speech-recognition", model="kmknair/wav2vec2-xlsr-tamil")
 p_ar = pipeline("automatic-speech-recognition", model="kmknair/wav2vec2-xlsr-arabic")
+p_en = pipeline("automatic-speech-recognition", mdoel="patrickvonplaten/hubert-xlarge-ls960-ft-4-gram")
 
 
 def transcribe_ta(audio_u, audio_m):        
@@ -39,6 +40,21 @@ def transcribe_ar_stream(audio, state=""):
     state += text + " " 
     return state, state
 
+def transcribe_en(audio_u, audio_m):    
+    text = ""
+    if audio_u is not None:
+        text += p_en(audio_u)["text"]
+    if audio_m is not None:
+        text += p_en(audio_m)["text"]
+    return text
+
+def transcribe_en_stream(audio, state=""):
+    time.sleep(2)
+    text = p_en(audio)["text"]
+    state += text + " " 
+    return state, state
+
+
 # transcribe Tamil stream
 ta_tr_stream_tab = gr.Interface(
     fn=transcribe_ta_stream,
@@ -64,6 +80,20 @@ ar_tr_stream_tab = gr.Interface(
     ],
     live=True)
 
+# transcribe English stream
+en_tr_stream_tab = gr.Interface(
+    fn=transcribe_en_stream,
+    inputs=[
+        gr.Audio(source="microphone", type="filepath", streaming=True),
+        "state"
+    ],
+    outputs=[
+    "textbox",
+    "state"
+    ],
+    live=True)
+
+
 # transcribe Tamil file
 ta_tr_file_tab = gr.Interface(
     fn=transcribe_ta,
@@ -82,14 +112,32 @@ ar_tr_file_tab = gr.Interface(
     ],
     outputs="text")    
 
+# transcribe English file
+en_tr_file_tab = gr.Interface(
+    fn=transcribe_en,
+    inputs=[    
+        gr.Audio(type="filepath"),
+        gr.Audio(source="microphone", type="filepath")
+    ],
+    outputs="text")       
+
 
 tabs = gr.TabbedInterface(
-    [ta_tr_stream_tab, ar_tr_stream_tab,ta_tr_file_tab, ar_tr_file_tab],
+    [   
+        ar_tr_stream_tab,
+        en_tr_stream_tab,
+        ta_tr_stream_tab,
+        ar_tr_file_tab,
+        en_tr_file_tab,
+        ta_tr_file_tab 
+    ],
     [
-        "Tamil Live Transcription",
         "Arabic Live Transcription",
-        "Tamil File Transcription",
+        "English Live Transcription",
+        "Tamil Live Transcription",
         "Arabic File Transcription",
+        "English File Transcription",
+        "Tamil File Transcription"
     ]
 )