From a11e1b8f3c5053b78014dcb9d1e9ac926418a9ed Mon Sep 17 00:00:00 2001 From: Niels Date: Sun, 7 Apr 2024 19:38:45 +0200 Subject: [PATCH 1/6] Add HF --- models/voicecraft.py | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/models/voicecraft.py b/models/voicecraft.py index ab3cf37..f090c66 100644 --- a/models/voicecraft.py +++ b/models/voicecraft.py @@ -18,6 +18,9 @@ from .modules.transformer import ( ) from .codebooks_patterns import DelayedPatternProvider +from huggingface_hub import PyTorchModelHubMixin + + def top_k_top_p_filtering( logits, top_k=0, top_p=1.0, filter_value=-float("Inf"), min_tokens_to_keep=1 ): @@ -82,7 +85,7 @@ def topk_sampling(logits, top_k=10, top_p=1.0, temperature=1.0): -class VoiceCraft(nn.Module): +class VoiceCraft(nn.Module, PyTorchModelHubMixin): def __init__(self, args): super().__init__() self.args = copy.copy(args) From 92b283c741d4110ff989e679405d805128299331 Mon Sep 17 00:00:00 2001 From: Niels Date: Sun, 7 Apr 2024 20:17:52 +0200 Subject: [PATCH 2/6] Add class --- models/voicecraft.py | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/models/voicecraft.py b/models/voicecraft.py index f090c66..8f87264 100644 --- a/models/voicecraft.py +++ b/models/voicecraft.py @@ -85,7 +85,7 @@ def topk_sampling(logits, top_k=10, top_p=1.0, temperature=1.0): -class VoiceCraft(nn.Module, PyTorchModelHubMixin): +class VoiceCraft(nn.Module): def __init__(self, args): super().__init__() self.args = copy.copy(args) @@ -1410,4 +1410,9 @@ class VoiceCraft(nn.Module, PyTorchModelHubMixin): res = res - int(self.args.n_special) flatten_gen = flatten_gen - int(self.args.n_special) - return res, flatten_gen[0].unsqueeze(0) \ No newline at end of file + return res, flatten_gen[0].unsqueeze(0) + + +class VoiceCraftHF(VoiceCraft, PyTorchModelHubMixin): + def __init__(self, config: dict): + super().__init__(config) \ No newline at end of file From 8ec653db9d5b05d186e7ccc28f9737499d9bbbdf Mon Sep 17 00:00:00 2001 From: Niels Date: Sun, 7 Apr 2024 20:21:39 +0200 Subject: [PATCH 3/6] Add class --- models/voicecraft.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/models/voicecraft.py b/models/voicecraft.py index 8f87264..8ea85ad 100644 --- a/models/voicecraft.py +++ b/models/voicecraft.py @@ -18,6 +18,7 @@ from .modules.transformer import ( ) from .codebooks_patterns import DelayedPatternProvider +from argparse import Namespace from huggingface_hub import PyTorchModelHubMixin @@ -1415,4 +1416,5 @@ class VoiceCraft(nn.Module): class VoiceCraftHF(VoiceCraft, PyTorchModelHubMixin): def __init__(self, config: dict): - super().__init__(config) \ No newline at end of file + args = Namespace(**config) + super().__init__(args) \ No newline at end of file From c44df99d499679c8e51f5ac0a7681a40ae76af50 Mon Sep 17 00:00:00 2001 From: Niels Date: Sun, 7 Apr 2024 20:34:27 +0200 Subject: [PATCH 4/6] Add tags --- models/voicecraft.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/models/voicecraft.py b/models/voicecraft.py index 8ea85ad..ee8e5a7 100644 --- a/models/voicecraft.py +++ b/models/voicecraft.py @@ -1414,7 +1414,9 @@ class VoiceCraft(nn.Module): return res, flatten_gen[0].unsqueeze(0) -class VoiceCraftHF(VoiceCraft, PyTorchModelHubMixin): +class VoiceCraftHF(VoiceCraft, PyTorchModelHubMixin, + repo_url="https://github.com/jasonppy/VoiceCraft", + tags=["Text-to-Speech, VoiceCraft"]): def __init__(self, config: dict): args = Namespace(**config) super().__init__(args) \ No newline at end of file From 0f79429b0d82c41190e8ad03bbe5b20112236417 Mon Sep 17 00:00:00 2001 From: Niels Date: Sun, 7 Apr 2024 20:36:35 +0200 Subject: [PATCH 5/6] Add tags --- models/voicecraft.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/models/voicecraft.py b/models/voicecraft.py index ee8e5a7..3c22105 100644 --- a/models/voicecraft.py +++ b/models/voicecraft.py @@ -1416,7 +1416,7 @@ class VoiceCraft(nn.Module): class VoiceCraftHF(VoiceCraft, PyTorchModelHubMixin, repo_url="https://github.com/jasonppy/VoiceCraft", - tags=["Text-to-Speech, VoiceCraft"]): + tags=["Text-to-Speech", "VoiceCraft"]): def __init__(self, config: dict): args = Namespace(**config) super().__init__(args) \ No newline at end of file From d363743fbd4fe58d54ffe218113a14c999d13a9c Mon Sep 17 00:00:00 2001 From: jason-on-salt-a40 Date: Mon, 8 Apr 2024 15:08:59 -0700 Subject: [PATCH 6/6] hf model --- README.md | 1 + models/voicecraft.py | 6 +++--- 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/README.md b/README.md index 8b7f984..a464833 100644 --- a/README.md +++ b/README.md @@ -95,6 +95,7 @@ pip install datasets==2.16.0 pip install torchmetrics==0.11.1 # install MFA for getting forced-alignment, this could take a few minutes conda install -c conda-forge montreal-forced-aligner=2.2.17 openfst=1.8.2 kaldi=5.5.1068 +pip install huggingface_hub # conda install pocl # above gives an warning for installing pocl, not sure if really need this # to run ipynb diff --git a/models/voicecraft.py b/models/voicecraft.py index 3c22105..cda380a 100644 --- a/models/voicecraft.py +++ b/models/voicecraft.py @@ -1414,9 +1414,9 @@ class VoiceCraft(nn.Module): return res, flatten_gen[0].unsqueeze(0) -class VoiceCraftHF(VoiceCraft, PyTorchModelHubMixin, - repo_url="https://github.com/jasonppy/VoiceCraft", - tags=["Text-to-Speech", "VoiceCraft"]): +class VoiceCraftHF(VoiceCraft, PyTorchModelHubMixin): + repo_url="https://github.com/jasonppy/VoiceCraft", + tags=["Text-to-Speech", "VoiceCraft"] def __init__(self, config: dict): args = Namespace(**config) super().__init__(args) \ No newline at end of file