ZenQin commited on
Commit
a53e350
1 Parent(s): 1d0d08a

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +100 -0
README.md CHANGED
@@ -1,3 +1,103 @@
1
  ---
2
  license: mit
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: mit
3
+ language:
4
+ - ko
5
+ pipeline_tag: text-to-speech
6
  ---
7
+
8
+ # MeloTTS
9
+
10
+ MeloTTS is a **high-quality multi-lingual** text-to-speech library by [MyShell.ai](https://myshell.ai). Supported languages include:
11
+
12
+
13
+ | Model card | Example |
14
+ | --- | --- |
15
+ | [English](https://huggingface.co/myshell-ai/MeloTTS-English-v2) (American) | [Link](https://myshell-public-repo-hosting.s3.amazonaws.com/myshellttsbase/examples/en/EN-US/speed_1.0/sent_000.wav) |
16
+ | [English](https://huggingface.co/myshell-ai/MeloTTS-English-v2) (British) | [Link](https://myshell-public-repo-hosting.s3.amazonaws.com/myshellttsbase/examples/en/EN-BR/speed_1.0/sent_000.wav) |
17
+ | [English](https://huggingface.co/myshell-ai/MeloTTS-English-v2) (Indian) | [Link](https://myshell-public-repo-hosting.s3.amazonaws.com/myshellttsbase/examples/en/EN_INDIA/speed_1.0/sent_000.wav) |
18
+ | [English](https://huggingface.co/myshell-ai/MeloTTS-English-v2) (Australian) | [Link](https://myshell-public-repo-hosting.s3.amazonaws.com/myshellttsbase/examples/en/EN-AU/speed_1.0/sent_000.wav) |
19
+ | [English](https://huggingface.co/myshell-ai/MeloTTS-English-v2) (Default) | [Link](https://myshell-public-repo-hosting.s3.amazonaws.com/myshellttsbase/examples/en/EN-Default/speed_1.0/sent_000.wav) |
20
+ | [Spanish](https://huggingface.co/myshell-ai/MeloTTS-Spanish) | [Link](https://myshell-public-repo-hosting.s3.amazonaws.com/myshellttsbase/examples/es/ES/speed_1.0/sent_000.wav) |
21
+ | [French](https://huggingface.co/myshell-ai/MeloTTS-French) | [Link](https://myshell-public-repo-hosting.s3.amazonaws.com/myshellttsbase/examples/fr/FR/speed_1.0/sent_000.wav) |
22
+ | [Chinese](https://huggingface.co/myshell-ai/MeloTTS-Chinese) (mix EN) | [Link](https://myshell-public-repo-hosting.s3.amazonaws.com/myshellttsbase/examples/zh/ZH/speed_1.0/sent_008.wav) |
23
+ | [Japanese](https://huggingface.co/myshell-ai/MeloTTS-Japanese) | [Link](https://myshell-public-repo-hosting.s3.amazonaws.com/myshellttsbase/examples/jp/JP/speed_1.0/sent_000.wav) |
24
+ | [Korean](https://huggingface.co/myshell-ai/MeloTTS-Korean/) | [Link](https://myshell-public-repo-hosting.s3.amazonaws.com/myshellttsbase/examples/kr/KR/speed_1.0/sent_000.wav) |
25
+
26
+ Some other features include:
27
+ - The Chinese speaker supports `mixed Chinese and English`.
28
+ - Fast enough for `CPU real-time inference`.
29
+
30
+
31
+ ## Usage
32
+
33
+ ### Without Installation
34
+
35
+ An unofficial [live demo](https://huggingface.co/spaces/mrfakename/MeloTTS) is hosted on Hugging Face Spaces.
36
+
37
+ #### Use it on MyShell
38
+
39
+ There are hundreds of TTS models on MyShell, much more than MeloTTS. See examples [here](https://github.com/myshell-ai/MeloTTS/blob/main/docs/quick_use.md#use-melotts-without-installation).
40
+ More can be found at the widget center of [MyShell.ai](https://app.myshell.ai/robot-workshop).
41
+
42
+ ### Install and Use Locally
43
+
44
+ Follow the installation steps [here](https://github.com/myshell-ai/MeloTTS/blob/main/docs/install.md#linux-and-macos-install) before using the following snippet:
45
+
46
+ ```python
47
+ from melo.api import TTS
48
+
49
+ # Speed is adjustable
50
+ speed = 1.0
51
+
52
+ # CPU is sufficient for real-time inference.
53
+ # You can set it manually to 'cpu' or 'cuda' or 'cuda:0' or 'mps'
54
+ device = 'auto' # Will automatically use GPU if available
55
+
56
+ # English
57
+ text = "Did you ever hear a folk tale about a giant turtle?"
58
+ model = TTS(language='EN_V2', device=device)
59
+ speaker_ids = model.hps.data.spk2id
60
+
61
+ # American accent
62
+ output_path = 'en-us.wav'
63
+ model.tts_to_file(text, speaker_ids['EN-US'], output_path, speed=speed)
64
+
65
+ # British accent
66
+ output_path = 'en-br.wav'
67
+ model.tts_to_file(text, speaker_ids['EN-BR'], output_path, speed=speed)
68
+
69
+ # Indian accent
70
+ output_path = 'en-india.wav'
71
+ model.tts_to_file(text, speaker_ids['EN_INDIA'], output_path, speed=speed)
72
+
73
+ # Australian accent
74
+ output_path = 'en-au.wav'
75
+ model.tts_to_file(text, speaker_ids['EN-AU'], output_path, speed=speed)
76
+
77
+ # Default accent
78
+ output_path = 'en-default.wav'
79
+ model.tts_to_file(text, speaker_ids['EN-Default'], output_path, speed=speed)
80
+
81
+ ```
82
+
83
+
84
+ ## Join the Community
85
+
86
+ **Open Source AI Grant**
87
+
88
+ We are actively sponsoring open-source AI projects. The sponsorship includes GPU resources, fundings and intellectual support (collaboration with top research labs). We welcome both reseach and engineering projects, as long as the open-source community needs them. Please contact [Zengyi Qin](https://www.qinzy.tech/) if you are interested.
89
+
90
+ **Contributing**
91
+
92
+ If you find this work useful, please consider contributing to the GitHub [repo](https://github.com/myshell-ai/MeloTTS).
93
+
94
+ - Many thanks to [@fakerybakery](https://github.com/fakerybakery) for adding the Web UI and CLI part.
95
+
96
+ ## License
97
+
98
+ This library is under MIT License, which means it is free for both commercial and non-commercial use.
99
+
100
+ ## Acknowledgements
101
+
102
+ This implementation is based on [TTS](https://github.com/coqui-ai/TTS), [VITS](https://github.com/jaywalnut310/vits), [VITS2](https://github.com/daniilrobnikov/vits2) and [Bert-VITS2](https://github.com/fishaudio/Bert-VITS2). We appreciate their awesome work.
103
+