Update app.py

app.py

@@ -1,5 +1,4 @@
 import gradio as gr
-from gradio_client import Client
 import requests
 import soundfile as sf
 import numpy as np
@@ -8,14 +7,11 @@ from pydub import AudioSegment
 import io
 
 # Define the Hugging Face Inference API URLs and headers
+ASR_API_URL = "https://api-inference.huggingface.co/models/Baghdad99/saad-speech-recognition-hausa-audio-to-text"
 TTS_API_URL = "https://api-inference.huggingface.co/models/Baghdad99/english_voice_tts"
 TRANSLATION_API_URL = "https://api-inference.huggingface.co/models/Baghdad99/saad-hausa-text-to-english-text"
 headers = {"Authorization": "Bearer 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"}
 
-
-# Define the Gradio client
-client = Client("https://baghdad99-baghdad99-saad-speech-recognition-haus-28684af.hf.space/--replicas/22eapbdds/")
-
 # Define the function to query the Hugging Face Inference API
 def query(api_url, payload):
     response = requests.post(api_url, headers=headers, json=payload)
@@ -26,8 +22,17 @@ def translate_speech(audio_file):
     print(f"Type of audio: {type(audio_file)}, Value of audio: {audio_file}")  # Debug line
 
     # Use the ASR pipeline to transcribe the audio
-    result = client.predict(audio_file.name, api_name="/predict")  # Change this line
-    transcription = result["text"]
+    with open(audio_file.name, "rb") as f:  # Change this line
+        data = f.read()
+    response = requests.post(ASR_API_URL, headers=headers, data=data)
+    output = response.json()
+
+    # Check if the output contains 'text'
+    if 'text' in output:
+        transcription = output["text"]
+    else:
+        print("The output does not contain 'text'")
+        return
 
     # Use the translation pipeline to translate the transcription
     translated_text = query(TRANSLATION_API_URL, {"inputs": transcription})