Try again

app.py

@@ -1,43 +1,58 @@
+import torch
+
 import gradio as gr
+import yt_dlp as youtube_dl
+from transformers import pipeline
+from transformers.pipelines.audio_utils import ffmpeg_read
+
+import tempfile
+import os
+
+MODEL_NAME = "dmatekenya/whisper-large-v3-chichewa"
+BATCH_SIZE = 8
+FILE_LIMIT_MB = 1000
+YT_LENGTH_LIMIT_S = 3600  # limit to 1 hour YouTube files
+
+device = 0 if torch.cuda.is_available() else "cpu"
+
+pipe = pipeline(
+    task="automatic-speech-recognition",
+    model=MODEL_NAME,
+    chunk_length_s=30,
+    device=device,
+)
+
+
+def transcribe(inputs, task):
+    if inputs is None:
+        raise gr.Error("No audio file submitted! Please upload or record an audio file before submitting your request.")
+
+    text = pipe(inputs, batch_size=BATCH_SIZE, generate_kwargs={"task": task}, return_timestamps=True)["text"]
+    return  text
+
+
+demo = gr.Blocks()
+
+file_transcribe = gr.Interface(
+    fn=transcribe,
+    inputs=[
+        gr.inputs.Audio(source="upload", type="filepath", optional=True, label="Audio file"),
+        gr.inputs.Radio(["transcribe", "translate"], label="Task", default="transcribe"),
+    ],
+    outputs="text",
+    layout="horizontal",
+    theme="huggingface",
+    title="Whisper Large V3: Transcribe Audio",
+    description=(
+        "Transcribe long-form microphone or audio inputs with the click of a button! Demo uses the OpenAI Whisper"
+        f" checkpoint [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) and 🤗 Transformers to transcribe audio files"
+        " of arbitrary length."
+    ),
+    allow_flagging="never",
+)
+
+with demo:
+    gr.TabbedInterface([file_transcribe], [ "Audio file"])
+
+demo.launch(enable_queue=True)
 
-import torch
-import librosa
-import json
-# Load model directly
-from transformers import pipeline, AutoProcessor, AutoModelForSpeechSeq2Seq
-
-pipe = pipeline("automatic-speech-recognition", model="dmatekenya/whisper-large-v3-chichewa")
-
-def transcribe(audio_file_mic=None, audio_file_upload=None, language="English (eng)"):
-    if audio_file_mic:
-        audio_file = audio_file_mic
-    elif audio_file_upload:
-        audio_file = audio_file_upload
-    else:
-        return "Please upload an audio file or record one"
-
-    # Make sure audio is 16kHz
-    # speech, sample_rate = librosa.load(audio_file)
-    # if sample_rate != 16000:
-    #     speech = librosa.resample(speech, orig_sr=sample_rate, target_sr=16000)
-
-    # Keep the same model in memory and simply switch out the language adapters by calling load_adapter() for the model and set_target_lang() for the tokenizer
-    # language_code = iso_codes[language]
-    # processor.tokenizer.set_target_lang(language_code)
-    # model.load_adapter(language_code)
-
-    result = pipe(audio_file)
-    return result["text"]
-
-
-description = ''''''
-
-iface = gr.Interface(fn=transcribe,
-                     inputs=[
-                         gr.Audio(source="microphone", type="filepath", label="Record Audio"),
-                         gr.Audio(source="upload", type="filepath", label="Upload Audio"),
-                         ],
-                     outputs=gr.Textbox(label="Transcription"),
-                     description=description
-                     )
-iface.launch()