app.py

import os
import gradio as gr
import glob

os.system("apt-get update -qq && apt-get install -qq libfluidsynth1 build-essential libasound2-dev libjack-dev")

# install t5x
os.system("git clone --branch=main https://github.com/google-research/t5x")
# pin T5X for python 3.7 compatibility
os.system("cd t5x; git reset --hard 2e05ad41778c25521738418de805757bf2e41e9e; cd ..")
os.system("mv t5x t5x_tmp; mv t5x_tmp/* .; rm -r t5x_tmp")
os.system("sed -i 's:jax\[tpu\]:jax:' setup.py")
os.system("python3 -m pip install -e .")

# install mt3
os.system("git clone --branch=main https://github.com/magenta/mt3")
os.system("mv mt3 mt3_tmp; mv mt3_tmp/* .; rm -r mt3_tmp")
os.system("python3 -m pip install -e .")

os.system("pip install jax[cuda]==0.3.15 -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html")

os.system("gsutil -q -m cp -r gs://mt3/checkpoints .")
os.system("gsutil -q -m cp gs://magentadata/soundfonts/SGM-v2.01-Sal-Guit-Bass-V1.3.sf2 .")

import functools
import os

import numpy as np
import tensorflow.compat.v2 as tf

import functools
import gin
import jax
import librosa
import note_seq
import seqio
import t5
import t5x

from mt3 import metrics_utils
from mt3 import models
from mt3 import network
from mt3 import note_sequences
from mt3 import preprocessors
from mt3 import spectrograms
from mt3 import vocabularies

import nest_asyncio
nest_asyncio.apply()

SAMPLE_RATE = 16000
SF2_PATH = 'SGM-v2.01-Sal-Guit-Bass-V1.3.sf2'

class InferenceModel(object):
  """Wrapper of T5X model for music transcription."""

  def __init__(self, checkpoint_path, model_type='mt3'):

    # Model Constants.
    if model_type == 'ismir2021':
      num_velocity_bins = 127
      self.encoding_spec = note_sequences.NoteEncodingSpec
      self.inputs_length = 512
    elif model_type == 'mt3':
      num_velocity_bins = 1
      self.encoding_spec = note_sequences.NoteEncodingWithTiesSpec
      self.inputs_length = 256
    else:
      raise ValueError('unknown model_type: %s' % model_type)

    gin_files = ['/home/user/app/mt3/gin/model.gin',
                 '/home/user/app/mt3/gin/mt3.gin']

    self.batch_size = 8
    self.outputs_length = 1024
    self.sequence_length = {'inputs': self.inputs_length, 
                            'targets': self.outputs_length}

    self.partitioner = t5x.partitioning.PjitPartitioner(
        model_parallel_submesh=(1, 1, 1, 1), num_partitions=1)

    # Build Codecs and Vocabularies.
    self.spectrogram_config = spectrograms.SpectrogramConfig()
    self.codec = vocabularies.build_codec(
        vocab_config=vocabularies.VocabularyConfig(
            num_velocity_bins=num_velocity_bins))
    self.vocabulary = vocabularies.vocabulary_from_codec(self.codec)
    self.output_features = {
        'inputs': seqio.ContinuousFeature(dtype=tf.float32, rank=2),
        'targets': seqio.Feature(vocabulary=self.vocabulary),
    }

    # Create a T5X model.
    self._parse_gin(gin_files)
    self.model = self._load_model()

    # Restore from checkpoint.
    self.restore_from_checkpoint(checkpoint_path)

  @property
  def input_shapes(self):
    return {
          'encoder_input_tokens': (self.batch_size, self.inputs_length),
          'decoder_input_tokens': (self.batch_size, self.outputs_length)
    }

  def _parse_gin(self, gin_files):
    """Parse gin files used to train the model."""
    gin_bindings = [
        'from __gin__ import dynamic_registration',
        'from mt3 import vocabularies',
        'VOCAB_CONFIG=@vocabularies.VocabularyConfig()',
        'vocabularies.VocabularyConfig.num_velocity_bins=%NUM_VELOCITY_BINS'
    ]
    with gin.unlock_config():
      gin.parse_config_files_and_bindings(
          gin_files, gin_bindings, finalize_config=False)

  def _load_model(self):
    """Load up a T5X `Model` after parsing training gin config."""
    model_config = gin.get_configurable(network.T5Config)()
    module = network.Transformer(config=model_config)
    return models.ContinuousInputsEncoderDecoderModel(
        module=module,
        input_vocabulary=self.output_features['inputs'].vocabulary,
        output_vocabulary=self.output_features['targets'].vocabulary,
        optimizer_def=t5x.adafactor.Adafactor(decay_rate=0.8, step_offset=0),
        input_depth=spectrograms.input_depth(self.spectrogram_config))


  def restore_from_checkpoint(self, checkpoint_path):
    """Restore training state from checkpoint, resets self._predict_fn()."""
    train_state_initializer = t5x.utils.TrainStateInitializer(
      optimizer_def=self.model.optimizer_def,
      init_fn=self.model.get_initial_variables,
      input_shapes=self.input_shapes,
      partitioner=self.partitioner)

    restore_checkpoint_cfg = t5x.utils.RestoreCheckpointConfig(
        path=checkpoint_path, mode='specific', dtype='float32')

    train_state_axes = train_state_initializer.train_state_axes
    self._predict_fn = self._get_predict_fn(train_state_axes)
    self._train_state = train_state_initializer.from_checkpoint_or_scratch(
        [restore_checkpoint_cfg], init_rng=jax.random.PRNGKey(0))

  @functools.lru_cache()
  def _get_predict_fn(self, train_state_axes):
    """Generate a partitioned prediction function for decoding."""
    def partial_predict_fn(params, batch, decode_rng):
      return self.model.predict_batch_with_aux(
          params, batch, decoder_params={'decode_rng': None})
    return self.partitioner.partition(
        partial_predict_fn,
        in_axis_resources=(
            train_state_axes.params,
            t5x.partitioning.PartitionSpec('data',), None),
        out_axis_resources=t5x.partitioning.PartitionSpec('data',)
    )

  def predict_tokens(self, batch, seed=0):
    """Predict tokens from preprocessed dataset batch."""
    prediction, _ = self._predict_fn(
        self._train_state.params, batch, jax.random.PRNGKey(seed))
    return self.vocabulary.decode_tf(prediction).numpy()

  def __call__(self, audio):
    """Infer note sequence from audio samples.
    
    Args:
      audio: 1-d numpy array of audio samples (16kHz) for a single example.
    Returns:
      A note_sequence of the transcribed audio.
    """
    ds = self.audio_to_dataset(audio)
    ds = self.preprocess(ds)

    model_ds = self.model.FEATURE_CONVERTER_CLS(pack=False)(
        ds, task_feature_lengths=self.sequence_length)
    model_ds = model_ds.batch(self.batch_size)

    inferences = (tokens for batch in model_ds.as_numpy_iterator()
                  for tokens in self.predict_tokens(batch))

    predictions = []
    for example, tokens in zip(ds.as_numpy_iterator(), inferences):
      predictions.append(self.postprocess(tokens, example))

    result = metrics_utils.event_predictions_to_ns(
        predictions, codec=self.codec, encoding_spec=self.encoding_spec)
    return result['est_ns']

  def audio_to_dataset(self, audio):
    """Create a TF Dataset of spectrograms from input audio."""
    frames, frame_times = self._audio_to_frames(audio)
    return tf.data.Dataset.from_tensors({
        'inputs': frames,
        'input_times': frame_times,
    })

  def _audio_to_frames(self, audio):
    """Compute spectrogram frames from audio."""
    frame_size = self.spectrogram_config.hop_width
    padding = [0, frame_size - len(audio) % frame_size]
    audio = np.pad(audio, padding, mode='constant')
    frames = spectrograms.split_audio(audio, self.spectrogram_config)
    num_frames = len(audio) // frame_size
    times = np.arange(num_frames) / self.spectrogram_config.frames_per_second
    return frames, times

  def preprocess(self, ds):
    pp_chain = [
        functools.partial(
            t5.data.preprocessors.split_tokens_to_inputs_length,
            sequence_length=self.sequence_length,
            output_features=self.output_features,
            feature_key='inputs',
            additional_feature_keys=['input_times']),
        # Cache occurs here during training.
        preprocessors.add_dummy_targets,
        functools.partial(
            preprocessors.compute_spectrograms,
            spectrogram_config=self.spectrogram_config)
    ]
    for pp in pp_chain:
      ds = pp(ds)
    return ds

  def postprocess(self, tokens, example):
    tokens = self._trim_eos(tokens)
    start_time = example['input_times'][0]
    # Round down to nearest symbolic token step.
    start_time -= start_time % (1 / self.codec.steps_per_second)
    return {
        'est_tokens': tokens,
        'start_time': start_time,
        # Internal MT3 code expects raw inputs, not used here.
        'raw_inputs': []
    }

  @staticmethod
  def _trim_eos(tokens):
    tokens = np.array(tokens, np.int32)
    if vocabularies.DECODED_EOS_ID in tokens:
      tokens = tokens[:np.argmax(tokens == vocabularies.DECODED_EOS_ID)]
    return tokens

inference_model = InferenceModel('/home/user/app/checkpoints/mt3/', 'mt3')

def inference(url):
  os.system(f"yt-dlp -x {url} -o 'audio.%(ext)s'")
  audio_file = glob.glob('audio.*')[0]
  with open(audio_file, 'rb') as f:
    data = f.read()
  audio = note_seq.audio_io.wav_data_to_samples_librosa(data, sample_rate=SAMPLE_RATE)
  est_ns = inference_model(audio)
  midi_file = f"./transcribed.mid"
  note_seq.sequence_proto_to_midi_file(est_ns, midi_file)
  return midi_file
  
title = "YouTube-to-MT3"
description = "Upload YouTube audio to MT3: Multi-Task Multitrack Music Transcription. Thanks to <a href=\"https://huggingface.co/spaces/akhaliq/MT3\">akhaliq</a> for the original <i>Spaces</i> implementation."

article = "<p style='text-align: center'><a href='https://arxiv.org/abs/2111.03017' target='_blank'>MT3: Multi-Task Multitrack Music Transcription</a> | <a href='https://github.com/magenta/mt3' target='_blank'>Github Repo</a></p>"

gr.Interface(
    inference, 
    gr.Textbox(label="Audio URL"),
    gr.outputs.File(label="Transcribed MIDI"),
    title=title,
    description=description,
    article=article,
    enable_queue=True
    ).launch()