"""Meta's XLS-R based speaker embedding.
- feature dimension: 768
- source: https://huggingface.co/docs/transformers/en/model_doc/wav2vec2#transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForPreTrainingOutput
"""
from typing import Optional

import torch
import librosa
import numpy as np
from transformers import AutoFeatureExtractor, AutoModelForPreTraining


class Wav2VecEmbedding:

    def __init__(self, ckpt: str = "facebook/wav2vec2-large-xlsr-53"):
        self.processor = AutoFeatureExtractor.from_pretrained(ckpt)
        self.model = AutoModelForPreTraining.from_pretrained(ckpt)
        self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
        self.model.to(self.device)
        self.model.eval()

    def get_speaker_embedding(self, wav: np.ndarray, sampling_rate: Optional[int] = None) -> np.ndarray:
        # audio file is decoded on the fly
        if sampling_rate != self.processor.sampling_rate:
            wav = librosa.resample(wav, orig_sr=sampling_rate, target_sr=self.processor.sampling_rate)
        inputs = self.processor(wav, sampling_rate=self.processor.sampling_rate, return_tensors="pt")
        with torch.no_grad():
            outputs = self.model(**{k: v.to(self.device) for k, v in inputs.items()})
        return outputs.projected_states.mean(1).cpu().numpy()[0]


class XLSR2BEmbedding(Wav2VecEmbedding):

    def __init__(self):
        super().__init__("facebook/wav2vec2-xls-r-2b")


class XLSR1BEmbedding(Wav2VecEmbedding):

    def __init__(self):
        super().__init__("facebook/wav2vec2-xls-r-1b")


class XLSR300MEmbedding(Wav2VecEmbedding):

    def __init__(self):
        super().__init__("facebook/wav2vec2-xls-r-300m")