Spaces:
Sleeping
Sleeping
File size: 1,606 Bytes
0c97faa db2ba20 2b1e81b d2e67c6 0c97faa ccadd27 db2ba20 d67a1b2 d2e67c6 0c97faa |
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 |
import gradio as gr
from huggingface_hub import InferenceClient
from transformers import pipeline
from scipy.io.wavfile import write as write_wav
AUDIO_FILE_PATH = "bark_generation.wav"
synthesizer = pipeline("text-to-speech", "suno/bark-small")
"""
For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
"""
client = InferenceClient("meta-llama/Meta-Llama-3-8B-Instruct")
with gr.Blocks() as demo:
chatbot = gr.Chatbot(type="messages")
audio_box = gr.Audio(autoplay=True)
msg = gr.Textbox(submit_btn=True)
clear = gr.Button("Clear")
def synthesize_audio(text):
speech = synthesizer(text, forward_params={"do_sample": True})
write_wav(AUDIO_FILE_PATH, rate=speech["sampling_rate"], data=speech["audio"])
def user(user_message, history: list):
return "", history + [{"role": "user", "content": user_message}]
def bot(history: list):
history.append({"role": "assistant", "content": ""})
for message in client.chat_completion(
history,
stream=True,
):
token = message.choices[0].delta.content
history[-1]["content"] += token
yield history, None
synthesize_audio(history[-1]["content"])
return history, AUDIO_FILE_PATH
msg.submit(user, [msg, chatbot], [msg, chatbot], queue=False).then(
bot, chatbot, [chatbot, audio_box]
)
clear.click(lambda: None, None, chatbot, queue=False)
if __name__ == "__main__":
demo.launch()
|