File size: 3,078 Bytes
9206300
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import torch
from inference.svs.base_svs_infer import BaseSVSInfer
from utils import load_ckpt
from utils.hparams import hparams
from modulesmodules.diff.shallow_diffusion_tts import GaussianDiffusion
from tasks.svs.diffsinger_task import DIFF_DECODERS

class DiffSingerCascadeInfer(BaseSVSInfer):
    def build_model(self):
        model = GaussianDiffusion(
            phone_encoder=self.ph_encoder,
            out_dims=hparams['audio_num_mel_bins'], denoise_fn=DIFF_DECODERS[hparams['diff_decoder_type']](hparams),
            timesteps=hparams['timesteps'],
            K_step=hparams['K_step'],
            loss_type=hparams['diff_loss_type'],
            spec_min=hparams['spec_min'], spec_max=hparams['spec_max'],
        )
        model.eval()
        load_ckpt(model, hparams['work_dir'], 'model')
        return model

    def forward_model(self, inp):
        sample = self.input_to_batch(inp)
        txt_tokens = sample['txt_tokens']  # [B, T_t]
        spk_id = sample.get('spk_ids')
        with torch.no_grad():
            output = self.model(txt_tokens, spk_id=spk_id, ref_mels=None, infer=True,
                                pitch_midi=sample['pitch_midi'], midi_dur=sample['midi_dur'],
                                is_slur=sample['is_slur'])
            mel_out = output['mel_out']  # [B, T,80]
            f0_pred = output['f0_denorm']
            wav_out = self.run_vocoder(mel_out, f0=f0_pred)
        wav_out = wav_out.cpu().numpy()
        return wav_out[0]


if __name__ == '__main__':
    inp = {
        'text': '小酒窝长睫毛AP是你最美的记号',
        'notes': 'C#4/Db4 | F#4/Gb4 | G#4/Ab4 | A#4/Bb4 F#4/Gb4 | F#4/Gb4 C#4/Db4 | C#4/Db4 | rest | C#4/Db4 | A#4/Bb4 | G#4/Ab4 | A#4/Bb4 | G#4/Ab4 | F4 | C#4/Db4',
        'notes_duration': '0.407140 | 0.376190 | 0.242180 | 0.509550 0.183420 | 0.315400 0.235020 | 0.361660 | 0.223070 | 0.377270 | 0.340550 | 0.299620 | 0.344510 | 0.283770 | 0.323390 | 0.360340',
        'input_type': 'word'
    }  # user input: Chinese characters
    c = {
        'text': '小酒窝长睫毛AP是你最美的记号',
        'ph_seq': 'x iao j iu w o ch ang ang j ie ie m ao AP sh i n i z ui m ei d e j i h ao',
        'note_seq': 'C#4/Db4 C#4/Db4 F#4/Gb4 F#4/Gb4 G#4/Ab4 G#4/Ab4 A#4/Bb4 A#4/Bb4 F#4/Gb4 F#4/Gb4 F#4/Gb4 C#4/Db4 C#4/Db4 C#4/Db4 rest C#4/Db4 C#4/Db4 A#4/Bb4 A#4/Bb4 G#4/Ab4 G#4/Ab4 A#4/Bb4 A#4/Bb4 G#4/Ab4 G#4/Ab4 F4 F4 C#4/Db4 C#4/Db4',
        'note_dur_seq': '0.407140 0.407140 0.376190 0.376190 0.242180 0.242180 0.509550 0.509550 0.183420 0.315400 0.315400 0.235020 0.361660 0.361660 0.223070 0.377270 0.377270 0.340550 0.340550 0.299620 0.299620 0.344510 0.344510 0.283770 0.283770 0.323390 0.323390 0.360340 0.360340',
        'is_slur_seq': '0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0',
        'input_type': 'phoneme'
    }  # input like Opencpop dataset.
    DiffSingerCascadeInfer.example_run(inp)

# # CUDA_VISIBLE_DEVICES=1 python inference/svs/ds_cascade.py --config egs/egs_bases/svs/midi/cascade/opencs/ds60_rel.yaml --exp_name 0303_opencpop_ds58_midi