multiple files

app.py

@@ -4,12 +4,10 @@ from asr import transcribe, ASR_LANGUAGES, model
 from tts import synthesize, TTS_EXAMPLES, TTS_LANGUAGES
 from lid import identify, LID_EXAMPLES
 
-
-
 mms_transcribe = gr.Interface(
-    fn=lambda audio, lang, transcription: transcribe(model, audio, lang, transcription),
+    fn=lambda audio_files, lang, transcription: [transcribe(model, audio, lang, transcription) for audio in audio_files],
     inputs=[
-        gr.Audio(),
+        gr.inputs.File(type="audio", multiple=True),  # Allow multiple audio files
         gr.Dropdown(
             [f"{k} ({v})" for k, v in ASR_LANGUAGES.items()],
             label="Language",
@@ -17,33 +15,12 @@ mms_transcribe = gr.Interface(
         ),
         gr.Textbox(label="Optional: Provide your own transcription"),
     ],
-    outputs="text",
+    outputs="text",  # Return a list of transcriptions
     title="Speech-to-text",
-    description=("Transcribe audio from a microphone or input file in your desired language."),
+    description=("Transcribe multiple audio files from a microphone or input files in your desired language."),
     allow_flagging="never",
 )
 
-#mms_synthesize = gr.Interface(
-    #fn=synthesize,
-    #inputs=[
-        #gr.Text(label="Input text"),
-        #gr.Dropdown(
-            #[f"{k} ({v})" for k, v in TTS_LANGUAGES.items()],
-            #label="Language",
-            #value="eng English",
-        #),
-        #gr.Slider(minimum=0.1, maximum=4.0, value=1.0, step=0.1, label="Speed"),
-    #],
-    #outputs=[
-       # gr.Audio(label="Generated Audio", type="numpy"),
-        #gr.Text(label="Filtered text after removing OOVs"),
-    #],
-    #examples=TTS_EXAMPLES,
-    #title="Text-to-speech",
-    #description=("Generate audio in your desired language from input text."),
-    #allow_flagging="never",
-#)
-
 mms_identify = gr.Interface(
     fn=identify,
     inputs=[