app.py

import gradio as gr

from fastai.vision.all import *

from fastaudio.core.all import *

matplotlib.rcParams['figure.dpi'] = 300

def get_x(df):
    return df.path
def get_y(df):
    return df.pattern
   
learn_removeSilence = load_learner('xresnet50_pitch3_removeSilence.pkl')

learn_plain = load_learner('xresnet50_pitch3.pkl')

labels = learn_removeSilence.dls.vocab

def process(Record, Upload, version):
    if version == 'remove silence':
        return predict(Record, Upload, learn_removeSilence)
    elif version == 'plain':
        return predict(Record, Upload, learn_plain)

def predict(Record, Upload, learn):
    if Upload: path = Upload
    else: path = Record
    spec,pred,pred_idx,probs = learn.predict(str(path), with_input=True)
    fig,ax = plt.subplots(figsize=(16,10))
    show_image(spec, ax=ax)
    ax.invert_yaxis()
    return [{labels[i]: float(probs[i]) for i in range(len(labels))}, fig]


title = "Japanese Pitch Accent Pattern Detector"

description = "This model will predict the pitch accent pattern of a word based on the recording of its pronunciation."

article="<p style='text-align: center'><a href='https://mizoru.github.io/blog/2021/12/25/Japanese-pitch.html' target='_blank'>How did I make this and what is it for?</a></p>"

examples = [['代わる.mp3'],['大丈夫な.mp3'],['熱くない.mp3'], ['あめー雨.mp3'], ['あめー飴.mp3']]

enable_queue=True

gr.Interface(fn=process,
inputs=[gr.inputs.Audio(source='microphone', type='filepath', optional=True),
gr.inputs.Audio(source='upload', type='filepath', optional=True),
gr.inputs.Radio(choices=['plain','remove silence'], type="value", default='remove silence', label='version')
],
outputs=  [gr.outputs.Label(num_top_classes=3), gr.outputs.Image(type="plot", label='Spectrogram')], title=title,description=description,article=article,examples=examples).launch(debug=True, enable_queue=enable_queue)