Spaces:
Sleeping
Sleeping
File size: 2,482 Bytes
2fcbfe2 2bf882b 2fcbfe2 cfba784 2fcbfe2 cfba784 2bf882b cfba784 fb4383a cfba784 2bf882b 2fcbfe2 cfba784 2bf882b cfba784 2fcbfe2 cfba784 2fcbfe2 2bf882b 2fcbfe2 2bf882b 2fcbfe2 cfba784 2bf882b 2fcbfe2 cfba784 2fcbfe2 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 | import gradio as gr
from huggingface_hub import InferenceClient
from transformers import pipeline
# Load LLM
client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
# Load Speech-to-Text (STT) model
stt_pipeline = pipeline("automatic-speech-recognition", model="openai/whisper-base")
# Load Text-to-Speech (TTS) model (using a public model without token requirements)
tts_pipeline = pipeline("text-to-speech", model="facebook/mms-tts-eng")
def respond(audio, message, history, system_message, max_tokens, temperature, top_p):
# Convert speech to text if audio input is provided
if audio is not None:
message = stt_pipeline(audio)["text"]
# Prepare conversation history
messages = [{"role": "system", "content": system_message}]
for user_msg, bot_msg in history:
if user_msg:
messages.append({"role": "user", "content": user_msg})
if bot_msg:
messages.append({"role": "assistant", "content": bot_msg})
messages.append({"role": "user", "content": message})
# Generate response from LLM
response = ""
for msg in client.chat_completion(
messages, max_tokens=max_tokens, stream=True, temperature=temperature, top_p=top_p
):
token = msg.choices[0].delta.content
response += token
# Convert chatbot response to speech
speech = tts_pipeline(response)
return history + [(message, response)], speech["audio"]
# Gradio Interface using Blocks
with gr.Blocks() as demo:
gr.Markdown("# 🎙️ Chatbot with Speech & Text")
with gr.Row():
audio_input = gr.Audio(type="filepath", label="🎤 Speak (or type below)")
text_input = gr.Textbox(label="💬 Or type your message")
chatbot = gr.Chatbot(label="Chat History")
with gr.Row():
system_msg = gr.Textbox(value="You are a friendly AI chatbot.", label="System Message")
max_tokens = gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max Tokens")
temperature = gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature")
top_p = gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p")
audio_output = gr.Audio(label="🔊 AI Response")
submit = gr.Button("Send")
submit.click(
respond,
inputs=[audio_input, text_input, chatbot, system_msg, max_tokens, temperature, top_p],
outputs=[chatbot, audio_output]
)
if __name__ == "__main__":
demo.launch()
|