Spaces:
Runtime error
Runtime error
File size: 1,788 Bytes
e130e80 bd27e22 e130e80 bd27e22 92f8ba8 e130e80 bd27e22 e130e80 bd27e22 e130e80 bd27e22 e130e80 92f8ba8 bd27e22 92f8ba8 bd27e22 92f8ba8 e130e80 515a989 e130e80 |
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 |
import gradio as gr
from transformers import pipeline
model_names = [
"juliensimon/wav2vec2-conformer-rel-pos-large-finetuned-speech-commands",
"MIT/ast-finetuned-speech-commands-v2",
]
def process(file, model_name):
p = pipeline("audio-classification", model=model_name)
pred = p(file)
return {x["label"]: x["score"] for x in pred}
# Gradio inputs
mic = gr.Audio(source="microphone", type="filepath", label="Speech input")
model_selection = gr.Dropdown(model_names, label="Model selection")
# Gradio outputs
labels = gr.Label(num_top_classes=3)
description = "This Space showcases two audio classification models fine-tuned on the speech_commands dataset:\n\n - wav2vec2-conformer: 97.2% accuracy, added in transformers 4.20.0.\n - audio-spectrogram-transformer: 98.12% accuracy, added in transformers 4.25.1.\n \n They can spot one of the following keywords: 'Yes', 'No', 'Up', 'Down', 'Left', 'Right', 'On', 'Off', 'Stop', 'Go', 'Zero', 'One', 'Two', 'Three', 'Four', 'Five', 'Six', 'Seven', 'Eight', 'Nine', 'Bed', 'Bird', 'Cat', 'Dog', 'Happy', 'House', 'Marvin', 'Sheila', 'Tree', 'Wow', 'Backward', 'Forward', 'Follow', 'Learn', 'Visual'."
iface = gr.Interface(
theme="huggingface",
description=description,
fn=process,
inputs=[mic, model_selection],
outputs=[labels],
examples=[
["backward16k.wav", "MIT/ast-finetuned-speech-commands-v2"],
["happy16k.wav", "MIT/ast-finetuned-speech-commands-v2"],
["marvin16k.wav", "MIT/ast-finetuned-speech-commands-v2"],
["seven16k.wav", "MIT/ast-finetuned-speech-commands-v2"],
["stop16k.wav", "MIT/ast-finetuned-speech-commands-v2"],
["up16k.wav", "MIT/ast-finetuned-speech-commands-v2"],
],
allow_flagging="never",
)
iface.launch()
|