Update app.py

app.py

@@ -11,7 +11,6 @@ from PIL import Image
 import spaces
 import torch
 from diffusers import DiffusionPipeline
-from huggingface_hub import Client
 
 DESCRIPTION = """# Playground v2.5"""
 if not torch.cuda.is_available():
@@ -27,14 +26,7 @@ device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
 
 NUM_IMAGES_PER_PROMPT = 1
 
-languages = {
-    "fon": "fon (Fon)",
-    "yo": "yo (Yoruba)",
-    "fr": "fra (French)",
-    "en": "eng (English)"
-}
-
-valid_languages = set(languages.keys())
+valid_languages = {'fon', 'fr', 'yo', 'en'}
 
 if torch.cuda.is_available():
     pipe = DiffusionPipeline.from_pretrained(
@@ -87,7 +79,7 @@ def translate_to_english(phrase, src_lang):
 
 @spaces.GPU(enable_queue=True)
 def generate(
-    audio,
+    phrase: str,
     input_lang: str,
     negative_prompt: str = "",
     use_negative_prompt: bool = False,
@@ -103,20 +95,10 @@ def generate(
     seed = int(randomize_seed_fn(seed, randomize_seed))
     generator = torch.Generator().manual_seed(seed)
 
-    client = Client("https://afrinetwork-speak.hf.space/")
-    language_name = languages[input_lang]
-    result = client.predict(
-        "Record from Mic",
-        audio,
-        language_name,
-        api_name="/predict"
-    )
-    transcript = result["transcription"]
-
     if input_lang != 'en':
-        prompt = translate_to_english(transcript, input_lang)
+        prompt = translate_to_english(phrase, input_lang)
     else:
-        prompt = transcript
+        prompt = phrase
 
     if not use_negative_prompt:
         negative_prompt = None  # type: ignore
@@ -161,8 +143,14 @@ with gr.Blocks(css=css) as demo:
     )
     with gr.Group():
         with gr.Row():
-            input_lang = gr.Dropdown(choices=list(languages.values()), value='eng (English)', label='Input Language')
-            audio = gr.Audio(source="microphone", type="filepath", label="Audio Input")
+            input_lang = gr.Dropdown(choices=list(valid_languages), value='en', label='Input Language')
+            prompt = gr.Text(
+                label="Prompt",
+                show_label=False,
+                max_lines=1,
+                placeholder="Enter your prompt",
+                container=False,
+            )
             run_button = gr.Button("Run", scale=0)
         result = gr.Gallery(label="Result", columns=NUM_IMAGES_PER_PROMPT, show_label=False)
     with gr.Accordion("Advanced options", open=False):
@@ -208,11 +196,12 @@ with gr.Blocks(css=css) as demo:
 
     gr.Examples(
         examples=examples,
-        inputs=[audio, input_lang],
+        inputs=[prompt, input_lang],
         outputs=[result, seed],
         fn=generate,
         cache_examples=CACHE_EXAMPLES,
     )
+
     use_negative_prompt.change(
         fn=lambda x: gr.update(visible=x),
         inputs=use_negative_prompt,
@@ -222,13 +211,13 @@ with gr.Blocks(css=css) as demo:
 
     gr.on(
         triggers=[
-            audio.upload,
+            prompt.submit,
             negative_prompt.submit,
             run_button.click,
         ],
         fn=generate,
         inputs=[
-            audio,
+            prompt,
             input_lang,
             negative_prompt,
             use_negative_prompt,
@@ -237,10 +226,10 @@ with gr.Blocks(css=css) as demo:
             height,
             guidance_scale,
             randomize_seed,
-        ],
-        outputs=[result, seed],
-        api_name="run",
-    )
+       ],
+       outputs=[result, seed],
+       api_name="run",
+   )
 
 if __name__ == "__main__":
-    demo.queue(max_size=20).launch()
+   demo.queue(max_size=20).launch()