Upload 3 files

README.md

@@ -0,0 +1,14 @@
+---
+title: EDGE TTS
+emoji: 🔉
+colorFrom: pink
+colorTo: green
+sdk: gradio
+sdk_version: 5.9.1
+app_file: app.py
+pinned: false
+license: creativeml-openrail-m
+short_description: Answer in speech
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,72 @@
+import os
+import re
+import gradio as gr
+import edge_tts
+import asyncio
+import time
+import tempfile
+from huggingface_hub import InferenceClient
+
+css= '''
+#important{
+  display: none;
+}
+'''
+DESCRIPTION = """## EDGE TTS
+"""
+
+client = InferenceClient("mistralai/Mixtral-8x7B-Instruct-v0.1")
+
+system_instructions = "[INST] Answers by 🔉, Keep conversation very short, clear, friendly and concise."
+
+async def generate(prompt):
+    generate_kwargs = dict(
+        temperature=0.6,
+        max_new_tokens=256,
+        top_p=0.95,
+        repetition_penalty=1,
+        do_sample=True,
+        seed=42,
+    )
+    formatted_prompt = system_instructions + prompt + "[/INST]"
+    stream = client.text_generation(
+        formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=True)
+    output = ""
+    for response in stream:
+        output += response.token.text
+
+    communicate = edge_tts.Communicate(output)
+    with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as tmp_file:
+        tmp_path = tmp_file.name
+        await communicate.save(tmp_path)
+    yield tmp_path
+
+with gr.Blocks(css=css) as demo:    
+    gr.Markdown(DESCRIPTION)
+    with gr.Row():
+        user_input = gr.Textbox(label="Prompt")
+        input_text = gr.Textbox(label="Input Text", elem_id="important")
+        output_audio = gr.Audio(label="Audio", type="filepath",
+                        interactive=False,
+                        autoplay=True,
+                        elem_classes="audio")
+    with gr.Row():
+        translate_btn = gr.Button("Response")
+        translate_btn.click(fn=generate, inputs=user_input,
+                            outputs=output_audio, api_name="translate")        
+    
+    # Add examples
+    gr.Examples(
+        examples=[
+            ["What is AI?"],
+            ["Add 2*3345"],
+            ["Describe Mt. Everest"]
+        ],
+        inputs=user_input,
+        outputs=output_audio,
+        fn=generate,
+        cache_examples=True
+    )
+
+if __name__ == "__main__":
+    demo.queue(max_size=20).launch()

requirements.txt

@@ -0,0 +1,8 @@
+edge-tts
+gradio
+asyncio
+transformers
+torch
+audiosegment
+scipy
+librosa