yoruba asr

app.py

@@ -0,0 +1,29 @@
+import gradio as gr
+import whisper
+import torch
+
+# Load the Whisper model
+model = whisper.load_model("base")  # or "small", "medium", "large"
+
+# Transcription function
+def transcribe(audio):
+    if audio is None:
+        return "No audio received."
+    
+    # Whisper expects a path to a file
+    audio_path = audio  # gradio gives a temporary file path
+    result = model.transcribe(audio_path)
+    return result["text"]
+
+# Gradio interface
+iface = gr.Interface(
+    fn=transcribe,
+    inputs=gr.Audio(source="microphone", type="filepath", label="🎙 Record your voice"),
+    outputs="text",
+    live=False,
+    title="🗣 Whisper Transcription",
+    description="Speak for a few seconds and get the transcription using OpenAI's Whisper model."
+)
+
+# Launch the app
+iface.launch()

packages.txt

@@ -0,0 +1 @@
+ffmpeg

requirements.txt

@@ -0,0 +1,8 @@
+transformers
+torch
+torchaudio
+fastapi
+uvicorn
+pydub
+python-multipart
+soundfile