from transformers import pipeline
import tempfile
import gradio as gr
from neon_tts_plugin_coqui import CoquiTTS

pipe = pipeline(model="Yuyang2022/yue")  # change to "your-username/the-name-you-picked"
LANGUAGES = list(CoquiTTS.langs.keys())
coquiTTS = CoquiTTS()

def audio_tts(audio, language: str):
    text = pipe(audio)["text"]
    with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as fp:
        coquiTTS.get_tts(text, fp, speaker = {"language" : language})
        return fp.name

inputs = [gr.Audio(source="microphone", type="filepath"), 
            gr.Dropdown(label="Language", LANGUAGES, value="en")]
outputs = gr.Audio(label="Output")

demo = gr.Interface(fn=audio_tts, inputs=inputs, outputs=outputs，
                title="translation-speeh to speech",
                    description="Realtime demo for speech translation.",)

demo.launch()