konverner commited on
Commit
b3ebc7c
β€’
1 Parent(s): 7865f10

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +9 -69
README.md CHANGED
@@ -1,69 +1,9 @@
1
- # Few-Shot Voice Cloning
2
-
3
- This repository is an implementation of the pipeline for few-short voice cloning based on SpeechT5 architecture introduced in [ SpeechT5: Unified-Modal Encoder-Decoder Pre-Training for Spoken Language Processing](https://arxiv.org/abs/2110.07205).
4
- It is able to clone a voice from 15-30 seconds of audio recording in English (another languages are planned).
5
-
6
- # Getting Started
7
-
8
- Clone repository
9
- ```angular2html
10
- git clone https://github.com/konverner/deep-voice-cloning.git
11
- ```
12
-
13
- Install the modules
14
- ```angular2html
15
- pip install .
16
- ```
17
-
18
- Run traning specifying arguments using config file `training_config.json` or the console command, for example
19
- ```angular2html
20
- python scripts/train.py --audio_path scripts/input/hank.mp3 --output_dir /content/deep-voice-cloning/models
21
- ```
22
- Resulting model will be saved in `output_dir` directory. It will be used in the next step.
23
-
24
- Run inference specifying arguments using config file `inference_config.json` or the console command, for example
25
- ```angular2html
26
- python scripts/cloning_inference.py --model_path "/content/deep-voice-cloning/models/microsoft_speecht5_tts_hank"\
27
- --input_text 'do the things, not because they are easy, but because they are hard'\
28
- --output_path "scripts/output/do_the_things.wav"
29
- ```
30
-
31
- Resulting audio file will be saved as `output_path` file.
32
-
33
- # Docker
34
-
35
- To build docker image:
36
-
37
- ```
38
- docker build -t deep-voice-cloning .
39
- ```
40
-
41
- To pull docker image from Hub:
42
-
43
- ```angular2html
44
- docker pull konverner/deep-voice-cloning:latest
45
- ```
46
-
47
- To run image in a container:
48
-
49
- ```
50
- docker run -it --entrypoint=/bin/bash konverner/deep-voice-cloning
51
- ```
52
-
53
- To run training in a container for example:
54
-
55
- ```
56
- python scripts/train.py --audio_path scripts/input/hank.mp3 --output_dir models
57
- ```
58
-
59
- To run inference in a container for example:
60
-
61
- ```
62
- python scripts/cloning_inference.py --model_path models/microsoft_speecht5_tts_hank --input_text "do the things, not because they are easy, but because they are hard" --output_path scripts/output/do_the_things.wav
63
- ```
64
-
65
-
66
- # Notebook Examples
67
-
68
- Example of using CLI for training and inference can be found in [notebook](https://github.com/konverner/deep-voice-cloning/blob/main/notebooks/CLI_Example.ipynb)
69
-
 
1
+ ---
2
+ license: openrail
3
+ title: Deep Voice Cloning
4
+ sdk: gradio
5
+ emoji: πŸŒ–
6
+ colorFrom: yellow
7
+ colorTo: purple
8
+ pinned: true
9
+ ---