app.py

import gradio as gr
import torch
import torchaudio
import torch.nn as nn
import torch.nn.functional as F

# Definición de la clase M5
class M5(nn.Module):
    def __init__(self, n_input=1, n_output=35, stride=16, n_channel=32):
        super().__init__()
        self.conv1 = nn.Conv1d(n_input, n_channel, kernel_size=80, stride=stride)
        self.bn1 = nn.BatchNorm1d(n_channel)
        self.pool1 = nn.MaxPool1d(4)
        self.conv2 = nn.Conv1d(n_channel, n_channel, kernel_size=3)
        self.bn2 = nn.BatchNorm1d(n_channel)
        self.pool2 = nn.MaxPool1d(4)
        self.conv3 = nn.Conv1d(n_channel, 2 * n_channel, kernel_size=3)
        self.bn3 = nn.BatchNorm1d(2 * n_channel)
        self.pool3 = nn.MaxPool1d(4)
        self.conv4 = nn.Conv1d(2 * n_channel, 2 * n_channel, kernel_size=3)
        self.bn4 = nn.BatchNorm1d(2 * n_channel)
        self.pool4 = nn.MaxPool1d(4)
        self.fc1 = nn.Linear(2 * n_channel, n_output)

    def forward(self, x):
        x = self.conv1(x)
        x = F.relu(self.bn1(x))
        x = self.pool1(x)
        x = self.conv2(x)
        x = F.relu(self.bn2(x))
        x = self.pool2(x)
        x = self.conv3(x)
        x = F.relu(self.bn3(x))
        x = self.pool3(x)
        x = self.conv4(x)
        x = F.relu(self.bn4(x))
        x = self.pool4(x)
        x = F.avg_pool1d(x, x.shape[-1])
        x = x.permute(0, 2, 1)
        x = self.fc1(x)
        return F.log_softmax(x, dim=2)

# Definición de etiquetas
labels = ['backward', 'bed', 'bird', 'cat', 'dog', 'down', 'eight', 'five', 'follow',
          'forward', 'four', 'go', 'happy', 'house', 'learn', 'left', 'marvin', 'nine',
          'no', 'off', 'on', 'one', 'right', 'seven', 'sheila', 'six', 'stop', 'three',
          'tree', 'two', 'up', 'visual', 'wow', 'yes', 'zero']

# Funciones auxiliares
def label_to_index(word):
    return torch.tensor(labels.index(word))

def index_to_label(index):
    return labels[index]

def get_likely_index(tensor):
    return tensor.argmax(dim=-1)

# Cargar el modelo
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model = M5()
model.load_state_dict(torch.load("modelo_entrenado.pth", map_location=device))
model.to(device)
model.eval()

# Definir la función de inferencia
def predict(audio):
    waveform, sample_rate = torchaudio.load(audio)
    transform = torchaudio.transforms.Resample(orig_freq=sample_rate, new_freq=8000).to(device)
    waveform = waveform.to(device)
    waveform = transform(waveform)

    with torch.no_grad():
        output = model(waveform.unsqueeze(0))
        tensor = get_likely_index(output)
        prediction = index_to_label(tensor.squeeze())
    return prediction

# Crear la interfaz de Gradio
iface = gr.Interface(
    fn=predict,
    inputs=gr.Audio(type="filepath"),
    outputs="text",
    title="Reconocimiento de comandos de voz",
    description="Graba un comando de voz y el modelo lo predecirá."
)

# Lanzar la interfaz
iface.launch(share=True)