nickthelegend
commited on
Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,100 @@
|
|
1 |
-
|
2 |
-
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Amir Singh Model - Indian "Über Eats" Typ Voice Clone
|
2 |
+
|
3 |
+
# Amir Singh Model - Indian "Über Eats" Typ Voice Clone
|
4 |
+
|
5 |
+
## Overview
|
6 |
+
|
7 |
+
The **Amir Singh Model** is a voice cloning model trained to mimic the voice of an "Indian Über Eats Typ". It uses RVC (Retrieval-based Voice Conversion) technology for efficient and accurate voice synthesis. The model was developed and trained on minimal data and optimized for quick deployment and use.
|
8 |
+
**He is built different**
|
9 |
+
## Key Features
|
10 |
+
|
11 |
+
* **Voice Type**: Indian Über Eats Typ (Amir Singh)
|
12 |
+
* **Training Data**: 5 minutes of audio data
|
13 |
+
* **Epochs**: 250 epochs
|
14 |
+
* **Segmentation & Training**:
|
15 |
+
* Data segmentation: 5 hours
|
16 |
+
* Training time: 1 hour
|
17 |
+
* **Hardware Used**:
|
18 |
+
* GPU: NVIDIA RTX 4060 TI (8GB VRAM)
|
19 |
+
* RAM: 24GB
|
20 |
+
|
21 |
+
## About RVC (Retrieval-based Voice Conversion)
|
22 |
+
|
23 |
+
RVC is a cutting-edge technology designed for voice conversion and cloning. It employs a retrieval-based approach that ensures the generated voice closely resembles the target voice with minimal artifacts. RVC is highly efficient, making it suitable for training with limited data while delivering high-quality results.
|
24 |
+
|
25 |
+
### Why RVC?
|
26 |
+
|
27 |
+
* **Low Data Requirement**: High-quality voice models can be created with as little as a few minutes of training data.
|
28 |
+
* **Fast Training**: Optimized for quick model training and deployment.
|
29 |
+
* **High Fidelity**: Produces realistic and natural-sounding voice outputs.
|
30 |
+
|
31 |
+
## Model Specifications
|
32 |
+
|
33 |
+
* **Input**: Audio samples for training (5 minutes)
|
34 |
+
* **Output**: Synthetic voice resembling "Amir Singh" with high accuracy
|
35 |
+
* **Performance**: Designed to work efficiently on systems with moderate hardware capabilities
|
36 |
+
|
37 |
+
## Usage
|
38 |
+
|
39 |
+
To use the Amir Singh Model:
|
40 |
+
|
41 |
+
1. Install the necessary dependencies, including RVC.
|
42 |
+
2. Load the trained model in your preferred framework or platform.
|
43 |
+
3. Input text or audio for voice conversion or synthesis.
|
44 |
+
4. Generate outputs that replicate the Amir Singh voice
|
45 |
+
|
46 |
+
Overview
|
47 |
+
|
48 |
+
The Amir Singh Model is a voice cloning model trained to mimic the voice of an "Indian Über Eats Typ". It uses RVC (Retrieval-based Voice Conversion) technology for efficient and accurate voice synthesis. The model was developed and trained on minimal data and optimized for quick deployment and use.
|
49 |
+
|
50 |
+
Key Features
|
51 |
+
|
52 |
+
Voice Type: Indian Über Eats Typ (Amir Singh)
|
53 |
+
|
54 |
+
Training Data: 5 minutes of audio data
|
55 |
+
|
56 |
+
Epochs: 250 epochs
|
57 |
+
|
58 |
+
Segmentation & Training:
|
59 |
+
|
60 |
+
Data segmentation: 5 hours
|
61 |
+
|
62 |
+
Training time: 1 hour
|
63 |
+
|
64 |
+
Hardware Used:
|
65 |
+
|
66 |
+
GPU: NVIDIA RTX 4060 TI (8GB VRAM)
|
67 |
+
|
68 |
+
RAM: 24GB
|
69 |
+
|
70 |
+
About RVC (Retrieval-based Voice Conversion)
|
71 |
+
|
72 |
+
RVC is a cutting-edge technology designed for voice conversion and cloning. It employs a retrieval-based approach that ensures the generated voice closely resembles the target voice with minimal artifacts. RVC is highly efficient, making it suitable for training with limited data while delivering high-quality results.
|
73 |
+
|
74 |
+
Why RVC?
|
75 |
+
|
76 |
+
Low Data Requirement: High-quality voice models can be created with as little as a few minutes of training data.
|
77 |
+
|
78 |
+
Fast Training: Optimized for quick model training and deployment.
|
79 |
+
|
80 |
+
High Fidelity: Produces realistic and natural-sounding voice outputs.
|
81 |
+
|
82 |
+
Model Specifications
|
83 |
+
|
84 |
+
Input: Audio samples for training (5 minutes)
|
85 |
+
|
86 |
+
Output: Synthetic voice resembling "Amir Singh" with high accuracy
|
87 |
+
|
88 |
+
Performance: Designed to work efficiently on systems with moderate hardware capabilities
|
89 |
+
|
90 |
+
Usage
|
91 |
+
|
92 |
+
To use the Amir Singh Model:
|
93 |
+
|
94 |
+
Install the necessary dependencies, including RVC.
|
95 |
+
|
96 |
+
Load the trained model in your preferred framework or platform.
|
97 |
+
|
98 |
+
Input text or audio for voice conversion or synthesis.
|
99 |
+
|
100 |
+
Generate outputs that replicate the Amir Singh voice.
|