tasks/audio.py

from fastapi import APIRouter
from datetime import datetime
from datasets import load_dataset
from sklearn.metrics import accuracy_score
import random
import os
import numpy as np
import librosa
import joblib

from .utils.evaluation import AudioEvaluationRequest
from .utils.emissions import tracker, clean_emissions_data, get_space_info

from dotenv import load_dotenv
load_dotenv()

router = APIRouter()

DESCRIPTION = "Random Baseline"
ROUTE = "/audio"

def create_spec(dataset, target_sampling_rate=3000):
    spectograms = []
    audio_length = int(36000/(12000/target_sampling_rate))
    for d in dataset:  
        audio_sample = librosa.resample(
            d["audio"]["array"], 
            orig_sr= d["audio"]["sampling_rate"], 
            target_sr=target_sampling_rate
        )
        
        if len(audio_sample) == 0:
            continue     
        if len(audio_sample) < audio_length:
            padding_needed = audio_length - len(audio_sample)
            repeats = (padding_needed // len(audio_sample)) + 1
            audio_sample = np.concatenate([audio_sample] + [audio_sample[:padding_needed]] * repeats)[:audio_length]
        elif len(audio_sample) > audio_length:
            audio_sample = audio_sample[:audio_length]
        
        rms = np.sqrt(np.mean(np.square(audio_sample)))
        scalar = 10 ** (-20 / 20) / (rms + 1e-8)
        
        mel = librosa.feature.melspectrogram(
            y=audio_sample*scalar,
            sr=12000,
            n_fft=2048,
            hop_length=1024,
            n_mels=12,
            power=2.0,
        )
        mel_db = librosa.power_to_db(mel, ref=np.max)  
        mel_db_normalized = (mel_db - mel_db.mean()) / (mel_db.std() + 1e-8)
        spectograms.append(mel_db_normalized.T.flatten())
        
    return np.stack(spectograms)


@router.post(ROUTE, tags=["Audio Task"],
             description=DESCRIPTION)
async def evaluate_audio(request: AudioEvaluationRequest):
    """
    Evaluate audio classification for rainforest sound detection.
    
    Current Model: Random Baseline
    - Makes random predictions from the label space (0-1)
    - Used as a baseline for comparison
    """
    # Get space info
    username, space_url = get_space_info()

    # Define the label mapping
    LABEL_MAPPING = {
        "chainsaw": 0,
        "environment": 1
    }
    # Load and prepare the dataset
    # Because the dataset is gated, we need to use the HF_TOKEN environment variable to authenticate
    dataset = load_dataset(request.dataset_name,token=os.getenv("HF_TOKEN"))
    
    # Split dataset
    train_test = dataset["train"].train_test_split(test_size=request.test_size, seed=request.test_seed)
    test_dataset = train_test["test"]
    
    # Start tracking emissions
    tracker.start()
    tracker.start_task("inference")
    
    test_spec = create_spec(test_dataset)
    H = np.load("H.npy")
    W_test = np.dot(test_spec, H)
    model = joblib.load('model.joblib')
    
    #--------------------------------------------------------------------------------------------
    # YOUR MODEL INFERENCE CODE HERE
    # Update the code below to replace the random baseline by your model inference within the inference pass where the energy consumption and emissions are tracked.
    #--------------------------------------------------------------------------------------------   
    
    # Make random predictions (placeholder for actual model inference)
    true_labels = test_dataset["label"]
    predictions = model.predict(W_test)
    #--------------------------------------------------------------------------------------------
    # YOUR MODEL INFERENCE STOPS HERE
    #--------------------------------------------------------------------------------------------   
    
    # Stop tracking emissions
    emissions_data = tracker.stop_task()
    
    # Calculate accuracy
    accuracy = accuracy_score(true_labels, predictions)
    
    # Prepare results dictionary
    results = {
        "username": username,
        "space_url": space_url,
        "submission_timestamp": datetime.now().isoformat(),
        "model_description": DESCRIPTION,
        "accuracy": float(accuracy),
        "energy_consumed_wh": emissions_data.energy_consumed * 1000,
        "emissions_gco2eq": emissions_data.emissions * 1000,
        "emissions_data": clean_emissions_data(emissions_data),
        "api_route": ROUTE,
        "dataset_config": {
            "dataset_name": request.dataset_name,
            "test_size": request.test_size,
            "test_seed": request.test_seed
        }
    }
    
    return results