Spaces:
Runtime error
Runtime error
File size: 2,084 Bytes
6886c22 cd87e9f 6886c22 cd87e9f 6886c22 cd87e9f 6886c22 21763f8 6886c22 058508d 743edd5 5d47fc0 058508d 05eee07 70432f2 |
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 |
import torch
import torch.nn.functional as F
from transformers import AutoConfig, Wav2Vec2FeatureExtractor
from src.models import Wav2Vec2ForSpeechClassification #imported from https://github.com/m3hrdadfi/soxan
import gradio as gr
import librosa
device = torch.device("cpu")
model_name_or_path = "harshit345/xlsr-wav2vec-speech-emotion-recognition"
config = AutoConfig.from_pretrained(model_name_or_path)
feature_extractor = Wav2Vec2FeatureExtractor.from_pretrained(model_name_or_path)
sampling_rate = feature_extractor.sampling_rate
model = Wav2Vec2ForSpeechClassification.from_pretrained(model_name_or_path)
#load input file and resample to 16kHz
def load_data(path):
speech, sampling_rate = librosa.load(path)
if len(speech.shape) > 1:
speech = speech[:,0] + speech[:,1]
if sampling_rate != 16000:
speech = librosa.resample(speech, sampling_rate,16000)
return speech
#modified version of predict function from https://github.com/m3hrdadfi/soxan
def inference(path):
speech = load_data(path)
inputs = feature_extractor(speech, return_tensors="pt").input_values
with torch.no_grad():
logits = model(inputs).logits
scores = F.softmax(logits, dim=1).detach().cpu().numpy()[0]
outputs = {config.id2label[i]: float(round(score,2)) for i, score in enumerate(scores)}
return outputs
inputs = gr.inputs.Audio(label="Input Audio", type="filepath", source="microphone")
outputs = gr.outputs.Label(type="confidences", label = "Output Scores")
title = "Wav2Vec2 Speech Emotion Recognition"
description = "This is a demo of the Wav2Vec2 Speech Emotion Recognition model. Record an audio file and the top emotions inferred will be displayed."
examples = ['data/heart.wav', 'data/happy26.wav', 'data/jm24.wav', 'data/newton.wav', 'data/speeding.wav']
article = "<a href = 'https://github.com/m3hrdadfi/soxan'> Wav2Vec2 Speech Classification Github Repository"
iface = gr.Interface(inference, inputs, outputs, title=title, description=description, article=article, theme="peach", examples=examples)
iface.launch(debug=True)
|