This seems to work for multi-speaker

glow-tts-jaywalnut-multispeaker/logs/base/config.json

@@ -0,0 +1,55 @@
+{
+  "train": {
+    "use_cuda": true,
+    "log_interval": 20,
+    "seed": 1234,
+    "epochs": 10000,
+    "learning_rate": 1e0,
+    "betas": [0.9, 0.98],
+    "eps": 1e-9,
+    "warmup_steps": 4000,
+    "scheduler": "noam",
+    "batch_size": 32,
+    "ddi": true,
+    "fp16_run": true
+  },
+  "data": {
+    "load_mel_from_disk": false,
+    "training_files":"/home/bt/datasets/voicecurator_training/training.csv",
+    "validation_files":"/home/bt/datasets/voicecurator_training/validation.csv",
+    "text_cleaners":["english_cleaners"],
+    "max_wav_value": 32768.0,
+    "sampling_rate": 22050,
+    "filter_length": 1024,
+    "hop_length": 256,
+    "win_length": 1024,
+    "n_mel_channels": 80,
+    "mel_fmin": 0.0,
+    "mel_fmax": 8000.0,
+    "add_noise": true,
+    "add_space": false,
+    "cmudict_path": "data/cmu_dictionary"
+  },
+  "model": {
+    "hidden_channels": 192,
+    "filter_channels": 768,
+    "filter_channels_dp": 256,
+    "kernel_size": 3,
+    "p_dropout": 0.1,
+    "n_blocks_dec": 12,
+    "n_layers_enc": 6,
+    "n_heads": 2,
+    "p_dropout_dec": 0.05,
+    "dilation_rate": 1,
+    "kernel_size_dec": 5,
+    "n_block_layers": 4,
+    "n_sqz": 2,
+    "prenet": true,
+    "mean_only": true,
+    "hidden_channels_enc": 192,
+    "hidden_channels_dec": 192,
+    "window_size": 4,
+    "n_speakers": 64,
+    "gin_channels": 256
+  }
+}

glow-tts-jaywalnut-multispeaker/requirements.txt

@@ -0,0 +1,16 @@
+torch==1.3.0
+librosa==0.7.1
+scipy==1.3.0
+PyYAML==5.2
+numpy==1.16.4
+
+# To fix warnings:
+numba==0.48.0
+
+# For monotonic_align
+# (cd monotonic_align; python setup.py build_ext --inplace)
+Cython==0.29.12
+
+# Other library needs
+Unidecode==1.1.1
+inflect==4.1.0

glow-tts-jaywalnut-multispeaker/wrap_glow_model.py

@@ -0,0 +1,103 @@
+#!/usr/bin/env python3
+# From: https://github.com/pytorch/pytorch/issues/20356#issuecomment-545572400
+import torch
+
+# Import glow-tts code
+import sys
+sys.path.append("./glow-tts-multispeaker")
+import utils
+import models
+import commons
+from text.symbols import symbols
+
+# Nvidia Apex
+#from apex import amp
+#from apex.parallel import DistributedDataParallel as DDP
+
+# Data
+import numpy as np
+from text import text_to_sequence, cmudict
+
+trace_model = False
+if sys.argv[1] == 'true':
+  trace_model = True
+elif sys.argv[1] == 'false':
+  trace_model = False
+else:
+  raise Exception('Scripting or tracing?')
+
+torch.set_grad_enabled(False)
+
+def load(model_checkpoint_path):
+  model_dir = "./logs/base/"
+  hps = utils.get_hparams_from_dir(model_dir)
+  print('Checkpoint path: {}'.format(model_checkpoint_path))
+
+  with torch.no_grad():
+    # NB(bt): This should fix the following:
+    # RuntimeError: Cannot insert a Tensor that requires grad as a constant.
+    model = models.FlowGenerator(
+        len(symbols),
+        out_channels=hps.data.n_mel_channels,
+        **hps.model) #.to("cuda")
+
+    utils.load_checkpoint(model_checkpoint_path, model)
+    model.decoder.store_inverse() # do not calcuate jacobians for fast decoding
+    _ = model.eval()
+
+  return (hps, model)
+
+
+#model_checkpoint_path = '/home/bt/models/glow-tts-jaywalnut/pretrained.pth'
+model_checkpoint_path = '/home/bt/models/glow-tts-jaywalnut-multispeaker/G_998.pth'
+
+hps, glow_model = load(model_checkpoint_path)
+output_filename = 'glow_multispeaker.torchjit'
+
+
+cmu_dict = cmudict.CMUDict(hps.data.cmudict_path)
+
+tst_stn = "This is a test of the machine learning model. How does this sound?"
+seq = text_to_sequence(tst_stn, ['english_cleaners'], cmu_dict)
+sequence = np.array(seq)[None, :] # NB: None adds a new axis.
+print("".join([symbols[c] for c in sequence[0]]))
+
+torched = torch.from_numpy(sequence)
+
+x_tst = torch.autograd.Variable(torched).long()
+x_tst_lengths = torch.tensor([x_tst.shape[1]])
+
+
+speaker_id = 1
+speaker_g = torch.IntTensor([int(speaker_id)]).long()
+
+print('\n tst_stn', len(tst_stn), tst_stn)
+print('\n seq', len(seq), seq)
+print('\n sequence', len(sequence), sequence.size, sequence)
+print('\n torched', len(torched), torched.shape, torched)
+
+print('\n x_tst', len(x_tst), x_tst)
+print('\n x_tst_lengths', len(x_tst_lengths), x_tst_lengths)
+
+print('\n speaker_g', len(speaker_g), speaker_g)
+
+
+# NB: Tracing evaluates the model on input and unrolls and hardcodes branching
+# and loops. Scripting allows these to remain by converting the entire program
+# to TorchScript. Scripting seems MUCH faster, is also harder get working.
+#trace_model = True # TODO: Try True.
+if trace_model:
+    print('Tracing model and saving as: {}'.format(output_filename))
+    #mel_file = '/home/bt/dev/voder/data/mels/LJ002-0320.mel'
+    #mel = torch.load(mel_file, map_location=torch.device('cpu'))
+    # NB: Getting `Tracing failed sanity checks! Graphs differed across invocations!`
+    inputs = (x_tst, x_tst_lengths, speaker_g)
+    #traced_script_module = torch.jit.trace(glow_model, inputs, check_trace=True)
+    traced_script_module = torch.jit.trace(glow_model, inputs, check_trace=False)
+    traced_script_module.save(output_filename)
+else:
+    print('Saving script model as: {}'.format(output_filename))
+    container = torch.jit.script(glow_model)
+    container.save(output_filename)
+
+