Commit
•
6a1c135
0
Parent(s):
Duplicate from suno/bark
Browse filesCo-authored-by: Georg Kucsko <georg-suno@users.noreply.huggingface.co>
- .gitattributes +34 -0
- README.md +14 -0
- app.py +146 -0
- requirements.txt +5 -0
.gitattributes
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
11 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
14 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
15 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
16 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
17 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
18 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
19 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
20 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
21 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
22 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
23 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
24 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
25 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
26 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
27 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
28 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
29 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
30 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
31 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
32 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
33 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
34 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
title: Bark
|
3 |
+
emoji: 🐶
|
4 |
+
colorFrom: pink
|
5 |
+
colorTo: blue
|
6 |
+
sdk: gradio
|
7 |
+
sdk_version: 3.24.1
|
8 |
+
app_file: app.py
|
9 |
+
pinned: false
|
10 |
+
license: cc-by-nc-4.0
|
11 |
+
duplicated_from: suno/bark
|
12 |
+
---
|
13 |
+
|
14 |
+
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
app.py
ADDED
@@ -0,0 +1,146 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import numpy as np
|
2 |
+
import gradio as gr
|
3 |
+
from bark import SAMPLE_RATE, generate_audio, preload_models
|
4 |
+
from bark.generation import SUPPORTED_LANGS
|
5 |
+
|
6 |
+
DEBUG_MODE = False
|
7 |
+
|
8 |
+
if not DEBUG_MODE:
|
9 |
+
_ = preload_models()
|
10 |
+
|
11 |
+
AVAILABLE_PROMPTS = ["Unconditional", "Announcer"]
|
12 |
+
PROMPT_LOOKUP = {}
|
13 |
+
for _, lang in SUPPORTED_LANGS:
|
14 |
+
for n in range(10):
|
15 |
+
label = f"Speaker {n} ({lang})"
|
16 |
+
AVAILABLE_PROMPTS.append(label)
|
17 |
+
PROMPT_LOOKUP[label] = f"{lang}_speaker_{n}"
|
18 |
+
PROMPT_LOOKUP["Unconditional"] = None
|
19 |
+
PROMPT_LOOKUP["Announcer"] = "announcer"
|
20 |
+
|
21 |
+
default_text = "Hello, my name is Suno. And, uh — and I like pizza. [laughs]\nBut I also have other interests such as playing tic tac toe."
|
22 |
+
|
23 |
+
title = "<div style='text-align:left'>🐶 Bark</div>"
|
24 |
+
|
25 |
+
description = """
|
26 |
+
<div>
|
27 |
+
<a style="display:inline-block" href='https://github.com/suno-ai/bark'><img src='https://img.shields.io/github/stars/suno-ai/bark?style=social' /></a>
|
28 |
+
<a style='display:inline-block' href='https://discord.gg/J2B2vsjKuE'><img src='https://dcbadge.vercel.app/api/server/J2B2vsjKuE?compact=true&style=flat' /></a>
|
29 |
+
<a style="display:inline-block; margin-left: 1em" href="https://huggingface.co/spaces/suno/bark?duplicate=true"><img src="https://img.shields.io/badge/-Duplicate%20Space%20to%20skip%20the%20queue-blue?labelColor=white&style=flat&logo=data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAQCAYAAAAf8/9hAAAAAXNSR0IArs4c6QAAAP5JREFUOE+lk7FqAkEURY+ltunEgFXS2sZGIbXfEPdLlnxJyDdYB62sbbUKpLbVNhyYFzbrrA74YJlh9r079973psed0cvUD4A+4HoCjsA85X0Dfn/RBLBgBDxnQPfAEJgBY+A9gALA4tcbamSzS4xq4FOQAJgCDwV2CPKV8tZAJcAjMMkUe1vX+U+SMhfAJEHasQIWmXNN3abzDwHUrgcRGmYcgKe0bxrblHEB4E/pndMazNpSZGcsZdBlYJcEL9Afo75molJyM2FxmPgmgPqlWNLGfwZGG6UiyEvLzHYDmoPkDDiNm9JR9uboiONcBXrpY1qmgs21x1QwyZcpvxt9NS09PlsPAAAAAElFTkSuQmCC&logoWidth=14" alt="Duplicate Space"></a>
|
30 |
+
</div>
|
31 |
+
Bark is a universal text-to-audio model created by [Suno](www.suno.ai), with code publicly available [here](https://github.com/suno-ai/bark). \
|
32 |
+
Bark can generate highly realistic, multilingual speech as well as other audio - including music, background noise and simple sound effects. \
|
33 |
+
This demo should be used for research purposes only. Commercial use is strictly prohibited. \
|
34 |
+
The model output is not censored and the authors do not endorse the opinions in the generated content. \
|
35 |
+
Use at your own risk.
|
36 |
+
"""
|
37 |
+
|
38 |
+
article = """
|
39 |
+
|
40 |
+
## 🌎 Foreign Language
|
41 |
+
|
42 |
+
Bark supports various languages out-of-the-box and automatically determines language from input text. \
|
43 |
+
When prompted with code-switched text, Bark will even attempt to employ the native accent for the respective languages in the same voice.
|
44 |
+
|
45 |
+
Try the prompt:
|
46 |
+
|
47 |
+
```
|
48 |
+
Buenos días Miguel. Tu colega piensa que tu alemán es extremadamente malo. But I suppose your english isn't terrible.
|
49 |
+
```
|
50 |
+
|
51 |
+
## 🤭 Non-Speech Sounds
|
52 |
+
|
53 |
+
Below is a list of some known non-speech sounds, but we are finding more every day. \
|
54 |
+
Please let us know if you find patterns that work particularly well on Discord!
|
55 |
+
|
56 |
+
* [laughter]
|
57 |
+
* [laughs]
|
58 |
+
* [sighs]
|
59 |
+
* [music]
|
60 |
+
* [gasps]
|
61 |
+
* [clears throat]
|
62 |
+
* — or ... for hesitations
|
63 |
+
* ♪ for song lyrics
|
64 |
+
* capitalization for emphasis of a word
|
65 |
+
* MAN/WOMAN: for bias towards speaker
|
66 |
+
|
67 |
+
Try the prompt:
|
68 |
+
|
69 |
+
```
|
70 |
+
" [clears throat] Hello, my name is Suno. And, uh — and I like pizza. [laughs] But I also have other interests such as... ♪ singing ♪."
|
71 |
+
```
|
72 |
+
|
73 |
+
## 🎶 Music
|
74 |
+
Bark can generate all types of audio, and, in principle, doesn't see a difference between speech and music. \
|
75 |
+
Sometimes Bark chooses to generate text as music, but you can help it out by adding music notes around your lyrics.
|
76 |
+
|
77 |
+
Try the prompt:
|
78 |
+
|
79 |
+
```
|
80 |
+
♪ In the jungle, the mighty jungle, the lion barks tonight ♪
|
81 |
+
```
|
82 |
+
|
83 |
+
## 🧬 Voice Cloning
|
84 |
+
|
85 |
+
Bark has the capability to fully clone voices - including tone, pitch, emotion and prosody. \
|
86 |
+
The model also attempts to preserve music, ambient noise, etc. from input audio. \
|
87 |
+
However, to mitigate misuse of this technology, we limit the audio history prompts to a limited set of Suno-provided, fully synthetic options to choose from.
|
88 |
+
|
89 |
+
## 👥 Speaker Prompts
|
90 |
+
|
91 |
+
You can provide certain speaker prompts such as NARRATOR, MAN, WOMAN, etc. \
|
92 |
+
Please note that these are not always respected, especially if a conflicting audio history prompt is given.
|
93 |
+
|
94 |
+
Try the prompt:
|
95 |
+
|
96 |
+
```
|
97 |
+
WOMAN: I would like an oatmilk latte please.
|
98 |
+
MAN: Wow, that's expensive!
|
99 |
+
```
|
100 |
+
|
101 |
+
## Details
|
102 |
+
|
103 |
+
Bark model by [Suno](https://suno.ai/), including official [code](https://github.com/suno-ai/bark) and model weights. \
|
104 |
+
Gradio demo supported by 🤗 Hugging Face. Bark is licensed under a non-commercial license: CC-BY 4.0 NC, see details on [GitHub](https://github.com/suno-ai/bark).
|
105 |
+
|
106 |
+
|
107 |
+
"""
|
108 |
+
|
109 |
+
examples = [
|
110 |
+
["Please surprise me and speak in whatever voice you enjoy. Vielen Dank und Gesundheit!", "Unconditional"],#, 0.7, 0.7],
|
111 |
+
["Hello, my name is Suno. And, uh — and I like pizza. [laughs] But I also have other interests such as playing tic tac toe.", "Speaker 1 (en)"],#, 0.7, 0.7],
|
112 |
+
["Buenos días Miguel. Tu colega piensa que tu alem��n es extremadamente malo. But I suppose your english isn't terrible.", "Speaker 0 (es)"],#, 0.7, 0.7],
|
113 |
+
]
|
114 |
+
|
115 |
+
|
116 |
+
def gen_tts(text, history_prompt):#, temp_semantic, temp_waveform):
|
117 |
+
history_prompt = PROMPT_LOOKUP[history_prompt]
|
118 |
+
if DEBUG_MODE:
|
119 |
+
audio_arr = np.zeros(SAMPLE_RATE)
|
120 |
+
else:
|
121 |
+
audio_arr = generate_audio(text, history_prompt=history_prompt)#, text_temp=temp_semantic, waveform_temp=temp_waveform)
|
122 |
+
audio_arr = (audio_arr * 32767).astype(np.int16)
|
123 |
+
return (SAMPLE_RATE, audio_arr)
|
124 |
+
|
125 |
+
iface = gr.Interface(
|
126 |
+
fn=gen_tts,
|
127 |
+
inputs=[
|
128 |
+
gr.Textbox(label="Input Text", lines=2, value=default_text),
|
129 |
+
gr.Dropdown(AVAILABLE_PROMPTS, value="Speaker 1 (en)", label="Acoustic Prompt"),
|
130 |
+
# gr.Slider(minimum=0, maximum=1, step=0.01, value=0.7, label="Temp 1", info="Gen. temperature of semantic tokens. (lower is more conservative, higher is more diverse)"),
|
131 |
+
# gr.Slider(minimum=0, maximum=1, step=0.01, value=0.7, label="Temp 2", info="Gen. temperature of waveform tokens. (lower is more conservative, higher is more diverse)"),
|
132 |
+
],
|
133 |
+
outputs=[
|
134 |
+
gr.Audio(label="Generated Audio", type="numpy"),
|
135 |
+
],
|
136 |
+
title=title,
|
137 |
+
description=description,
|
138 |
+
article=article,
|
139 |
+
examples=examples,
|
140 |
+
cache_examples=False,
|
141 |
+
)
|
142 |
+
|
143 |
+
with gr.Group(elem_id="share-btn-container", visible=False):
|
144 |
+
share_button = gr.Button("Share to community", elem_id="share-btn")
|
145 |
+
|
146 |
+
iface.launch(enable_queue=True)
|
requirements.txt
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
git+https://github.com/suno-ai/bark.git
|
2 |
+
https://download.pytorch.org/whl/nightly/pytorch_triton-2.1.0%2B46672772b4-cp38-cp38-linux_x86_64.whl
|
3 |
+
https://download.pytorch.org/whl/nightly/cu117/torch-2.1.0.dev20230413%2Bcu117-cp38-cp38-linux_x86_64.whl
|
4 |
+
https://download.pytorch.org/whl/nightly/cu117/torchvision-0.16.0.dev20230413%2Bcu117-cp38-cp38-linux_x86_64.whl
|
5 |
+
https://download.pytorch.org/whl/nightly/cu117/torchaudio-2.1.0.dev20230413%2Bcu117-cp38-cp38-linux_x86_64.whl
|