File size: 4,656 Bytes
ec127c3
13efc84
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
import note_seq
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("TristanBehrens/js-fakes-4bars")
model = AutoModelForCausalLM.from_pretrained("TristanBehrens/js-fakes-4bars")

NOTE_LENGTH_16TH_120BPM = 0.25 * 60 / 120
BAR_LENGTH_120BPM = 4.0 * 60 / 120
SAMPLE_RATE=44100

def token_sequence_to_note_sequence(token_sequence, use_program=True, use_drums=True, instrument_mapper=None, only_piano=False):
    if isinstance(token_sequence, str):
        token_sequence = token_sequence.split()
    note_sequence = empty_note_sequence()

    # Render all notes.
    current_program = 1
    current_is_drum = False
    current_instrument = 0
    track_count = 0
    for token_index, token in enumerate(token_sequence):

        if token == "PIECE_START":
            pass
        elif token == "PIECE_END":
            print("The end.")
            break
        elif token == "TRACK_START":
            current_bar_index = 0
            track_count += 1
            pass
        elif token == "TRACK_END":
            pass
        elif token == "KEYS_START":
            pass
        elif token == "KEYS_END":
            pass
        elif token.startswith("KEY="):
            pass
        elif token.startswith("INST"):
            instrument = token.split("=")[-1]
            if instrument != "DRUMS" and use_program:
                if instrument_mapper is not None:
                    if instrument in instrument_mapper:
                        instrument = instrument_mapper[instrument]
                current_program = int(instrument)
                current_instrument = track_count
                current_is_drum = False
            if instrument == "DRUMS" and use_drums:
                current_instrument = 0
                current_program = 0
                current_is_drum = True
        elif token == "BAR_START":
            current_time = current_bar_index * BAR_LENGTH_120BPM
            current_notes = {}
        elif token == "BAR_END":
            current_bar_index += 1
            pass
        elif token.startswith("NOTE_ON"):
            pitch = int(token.split("=")[-1])
            note = note_sequence.notes.add()
            note.start_time = current_time
            note.end_time = current_time + 4 * NOTE_LENGTH_16TH_120BPM
            note.pitch = pitch
            note.instrument = current_instrument
            note.program = current_program
            note.velocity = 80
            note.is_drum = current_is_drum
            current_notes[pitch] = note
        elif token.startswith("NOTE_OFF"):
            pitch = int(token.split("=")[-1])
            if pitch in current_notes:
                note = current_notes[pitch]
                note.end_time = current_time
        elif token.startswith("TIME_DELTA"):
            delta = float(token.split("=")[-1]) * NOTE_LENGTH_16TH_120BPM
            current_time += delta
        elif token.startswith("DENSITY="):
            pass
        elif token == "[PAD]":
            pass
        else:
            #print(f"Ignored token {token}.")
            pass

    # Make the instruments right.
    instruments_drums = []
    for note in note_sequence.notes:
        pair = [note.program, note.is_drum]
        if pair not in instruments_drums:
            instruments_drums += [pair]
        note.instrument = instruments_drums.index(pair)

    if only_piano:
        for note in note_sequence.notes:
            if not note.is_drum:
                note.instrument = 0
                note.program = 0

    return note_sequence

def empty_note_sequence(qpm=120.0, total_time=0.0):
    note_sequence = note_seq.protobuf.music_pb2.NoteSequence()
    note_sequence.tempos.add().qpm = qpm
    note_sequence.ticks_per_quarter = note_seq.constants.STANDARD_PPQ
    note_sequence.total_time = total_time
    return note_sequence

def process(text):
  input_ids = tokenizer.encode(text, return_tensors="pt")
  generated_ids = model.generate(input_ids, max_length=500)
  generated_sequence = tokenizer.decode(generated_ids[0])
  print(generated_sequence)

  # Convert text of notes to audio
  note_sequence = token_sequence_to_note_sequence(generated_sequence)
  synth = note_seq.midi_synth.synthesize
  array_of_floats = synth(note_sequence, sample_rate=SAMPLE_RATE)
  return SAMPLE_RATE, array_of_floats
  
title = "Music generation with GPT-2"

iface = gr.Interface(
  fn=process, 
  inputs=[gr.inputs.Textbox(placeholder="PIECE_START")],
  outputs='audio',
  title=title,
  examples=[["PIECE_START"], ["PIECE_START STYLE=JSFAKES GENRE=JSFAKES TRACK_START INST=48 BAR_START NOTE_ON=61"]]
)

iface.launch(debug=True)