• Stars
    star
    184
  • Rank 209,187 (Top 5 %)
  • Language
  • Created over 6 years ago
  • Updated over 6 years ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

Making a TTS model with 1 minute of speech samples within 10 minutes

Making a TTS model with 1 minute of speech samples within 10 minutes

Seeing my implementaions of Tacotron and DCTTS, many people have asked me "How large speech dataset is needed for neural TTS?" or "Can you make a TTS model with X hour(s)/minute(s) of training data?" I'm fully aware of the importance of those questions. When you plan a service using TTS, it is not always likely to get lots of speech samples. I would like to give an answer. I really do. But unfortunately I have no answer. The only thing I know is that I could train a model successfully with five hours of speech samples I extracted from Kate Winslet's audiobook. I haven't tried less data than that. I could try it, but I actually I have a better idea. Since I have a decent model trained with the LJ Speech Dataset for several days, why don't I use it? After all, we all have different voices, but the way we speak English is not totally different.

In the above two repos, I trained TTS models using all the speech samples of my two favorite celebrities, Nick Offerman and Kate Winslet, from scratch. This time, I use only one minute of the speech samples. The following are the synthesized samples after 10 minutes of fine-tuning training. Do you think they sound like them?

Additionally, I collected 10 speech samples of Modern Family celebrities from YouTube, and generated their voice, training on those sample.

Check here to see the model details, source code and the pretrained model which served as a seed.

More Repositories

1

transformer

A TensorFlow Implementation of the Transformer: Attention Is All You Need
Python
4,237
star
2

nlp_tasks

Natural Language Processing Tasks and References
3,017
star
3

wordvectors

Pre-trained word vectors of 30+ languages
Python
2,205
star
4

tacotron

A TensorFlow Implementation of Tacotron: A Fully End-to-End Text-To-Speech Synthesis Model
Python
1,827
star
5

numpy_exercises

Numpy exercises.
Python
1,687
star
6

dc_tts

A TensorFlow Implementation of DC-TTS: yet another text-to-speech model
Python
1,159
star
7

sudoku

Can Neural Networks Crack Sudoku?
Python
823
star
8

g2p

g2p: English Grapheme To Phoneme Conversion
Python
783
star
9

tensorflow-exercises

TensorFlow Exercises - focusing on the comparison with NumPy.
Python
535
star
10

css10

CSS10: A Collection of Single Speaker Speech Datasets for 10 Languages
HTML
457
star
11

deepvoice3

Tensorflow Implementation of Deep Voice 3
Python
452
star
12

neural_chinese_transliterator

Can CNNs transliterate Pinyin into Chinese characters correctly?
Python
334
star
13

pytorch_exercises

Jupyter Notebook
313
star
14

bert_ner

Ner with Bert
Python
281
star
15

nlp_made_easy

Explains nlp building blocks in a simple manner.
Jupyter Notebook
251
star
16

word_prediction

Word Prediction using Convolutional Neural Networks
Python
251
star
17

g2pC

g2pC: A Context-aware Grapheme-to-Phoneme Conversion module for Chinese
Python
235
star
18

g2pK

g2pK: g2p module for Korean
Python
230
star
19

expressive_tacotron

Tensorflow Implementation of Expressive Tacotron
Python
198
star
20

neural_japanese_transliterator

Can neural networks transliterate Romaji into Japanese correctly?
Python
173
star
21

tacotron_asr

Speech Recognition Using Tacotron
Python
165
star
22

quasi-rnn

Character-level Neural Translation using Quasi-RNNs
Python
134
star
23

label_smoothing

Corrupted labels and label smoothing
Jupyter Notebook
128
star
24

name2nat

name2nat: a Python package for nationality prediction from a name
Python
98
star
25

bert-token-embeddings

Jupyter Notebook
97
star
26

cross_vc

Cross-lingual Voice Conversion
Python
95
star
27

mtp

Multi-lingual Text Processing
95
star
28

pron_dictionaries

pronunciation dictionaries for multiple languages
Python
83
star
29

msg_reply

a simple message reply suggestion system
Python
79
star
30

word_ordering

Can neural networks order a scramble of words correctly?
Python
73
star
31

kss

Python
71
star
32

neural_tokenizer

Tokenize English sentences using neural networks.
Python
64
star
33

bytenet_translation

A TensorFlow Implementation of Machine Translation In Neural Machine Translation in Linear Time
Python
60
star
34

KoParadigm

KoParadigm: Korean Inflectional Paradigm Generator
Python
54
star
35

specAugment

Tensor2tensor experiment with SpecAugment
Python
47
star
36

vq-vae

A Tensorflow Implementation of VQ-VAE Speaker Conversion
Python
43
star
37

lm_finetuning

Language Model Fine-tuning for Moby Dick
Python
42
star
38

texture_generation

An Implementation of 'Texture Synthesis Using Convolutional Neural Networks' with Kylberg Texture Dataset
Python
33
star
39

cjk_trans

Pre-trained Machine Translation Models of Korean from/to ECJ
29
star
40

h2h_converter

Convert Sino-Korean words written in Hangul to Chinese characters, which is called hanja in Korean, using neural networks
Python
28
star
41

integer_sequence_learning

RNN Approaches to Integer Sequence Learning--the famous Kaggle competition
Python
27
star
42

up_and_running_with_Tensorflow

A simple tutorial of TensorFlow + TensorFlow / NumPy exercises
Jupyter Notebook
13
star
43

neurobind

Yet Another Model Using Neural Networks for Predicting Binding Preferences of for Test DNA Sequences
Python
11
star
44

kollocate

Collocation Search of Korean
Python
10
star
45

kyubyong

9
star
46

WhereAmI

Where Am I? - If you want to meet me.
5
star
47

spam_detection

Spam Dectection Under Semi-supervised settings
5
star
48

helo_word

A Neural Grammatical Error Correction System Built On Better Pre-training and Sequential Transfer Learning
Python
2
star