|
import torchaudio |
|
from speechbrain.pretrained import EncoderClassifier |
|
import gradio as gr |
|
|
|
classifier = EncoderClassifier.from_hparams(source="speechbrain/lang-id-commonlanguage_ecapa", savedir="pretrained_models/lang-id-commonlanguage_ecapa") |
|
def speechbrain(aud): |
|
out_prob, score, index, text_lab = classifier.classify_file(aud.name) |
|
return text_lab[0] |
|
|
|
inputs = gr.inputs.Audio(label="Input Audio", type="file") |
|
outputs = "text" |
|
title = "Speechbrain Audio Classification" |
|
description = "Gradio demo for Audio Classification with SpeechBrain. To use it, simply upload your audio, or click one of the examples to load them. Read more at the links below." |
|
article = "<p style='text-align: center'><a href='https://arxiv.org/abs/2005.07143' target='_blank'>ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verification</a> | <a href='https://github.com/speechbrain/speechbrain' target='_blank'>Github Repo</a></p>" |
|
examples = [ |
|
['samples_audio_samples_example_fr.wav'] |
|
] |
|
gr.Interface(speechbrain, inputs, outputs, title=title, description=description, article=article, examples=examples).launch() |