address https://github.com/lucidrains/audiolm-pytorch/issues/76 and give thanks to Andrey

Update: AudioLM was essentially used to 'solve' music generation in the new <a href="https://github.com/lucidrains/audiolm-pytorch">MusicLM</a>

## Appreciation

- <a href="https://stability.ai/">Stability.ai</a> for the generous sponsorship to work and open source cutting edge artificial intelligence research

- <a href="https://huggingface.co/">🤗 Huggingface</a> for their amazing accelerate and transformers libraries

- <a href="https://ai.facebook.com/">MetaAI</a> for <a href="https://github.com/facebookresearch/fairseq">Fairseq</a> and the liberal license

- <a href="https://github.com/eonglints">@eonglints</a> for offering his professional advice and expertise as well as pull requests!

- <a href="https://github.com/djqualia">@djqualia</a>, <a href="https://github.com/yigityu">@yigityu</a>, <a href="https://github.com/inspirit">@inspirit</a>, and <a href="https://github.com/BlackFox1197">@BlackFox1197</a> for helping with the debugging of soundstream

- <a href="https://github.com/zhvng">Allen</a> for catching and fixing some bugs!

- <a href="https://github.com/AndreyBocharnikov">Andrey</a> for identifying a missing loss in soundstream and guiding me through the proper mel spectrogram hyperparameters

## Install

```bash

```

## Appreciation

- <a href="https://stability.ai/">Stability.ai</a> for the generous sponsorship to work and open source cutting edge artificial intelligence research

- <a href="https://huggingface.co/">🤗 Huggingface</a> for their amazing accelerate and transformers libraries

- <a href="https://ai.facebook.com/">MetaAI</a> for <a href="https://github.com/facebookresearch/fairseq">Fairseq</a> and the liberal license

- <a href="https://github.com/eonglints">@eonglints</a> for offering his professional advice and expertise as well as pull requests!

- <a href="https://github.com/djqualia">@djqualia</a>, <a href="https://github.com/yigityu">@yigityu</a>, <a href="https://github.com/inspirit">@inspirit</a>, and <a href="https://github.com/BlackFox1197">@BlackFox1197</a> for helping with the debugging of soundstream

- <a href="https://github.com/zhvng">Allen</a> for catching and fixing some bugs!

## Todo

- [x] complete CoarseTransformer

    return inner

# tensor helpers

def l2norm(t, dim = -1):

    return F.normalize(t, dim = dim)

# gan losses

def log(t, eps = 1e-20):

        self.mel_spec_transforms = nn.ModuleList([])

        self.mel_spec_recon_alphas = []

        max_win_length = 2 ** max(multi_spectral_window_powers_of_two)

        for powers in multi_spectral_window_powers_of_two:

            win_length = 2 ** powers

            alpha = (win_length / 2) ** 0.5

            melspec_transform = T.MelSpectrogram(

                sample_rate = target_sample_hz,

                n_fft = max_win_length,

                n_fft = win_length,

                win_length = win_length,

                hop_length = win_length // 4,

                n_mels = num_mel_bins

        for mel_transform, alpha in zip(self.mel_spec_transforms, self.mel_spec_recon_alphas):

            orig_mel, recon_mel = map(mel_transform, (orig_x, recon_x))

            log_orig_mel, log_recon_mel = map(log, (orig_mel, recon_mel))

            multi_spectral_recon_loss = multi_spectral_recon_loss + (orig_mel - recon_mel).abs().sum() + alpha * ((log_orig_mel - log_recon_mel) ** 2).sum()

            multi_spectral_recon_loss = multi_spectral_recon_loss + (orig_mel - recon_mel).abs().sum() + alpha * l2norm(log_orig_mel - log_recon_mel, dim = -2).sum()

        # adversarial loss

setup(

  name = 'audiolm-pytorch',

  packages = find_packages(exclude=[]),

  version = '0.11.1',

  version = '0.11.2',

  license='MIT',

  description = 'AudioLM - Language Modeling Approach to Audio Generation from Google Research - Pytorch',

  author = 'Phil Wang',