File size: 874 Bytes
7fed7ad
 
 
 
 
df6f32c
 
7fed7ad
 
 
c53dbfb
babd1b7
5b915ca
 
7fed7ad
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
---
license: mit
language:
- en
pipeline_tag: text-to-speech
library_name: vui

---
# vui

[DEMO](https://fluxions.ai)

https://github.com/fluxions-ai/vui

Small Conversational speech models that can run on device

# Installation

```sh
uv pip install -e .
```

# Demo

```sh
python demo.py
````

# Models

Vui.BASE is base checkpoint trained on 40k hours of audio conversations
Vui.ABRAHAM is a single speaker model that can reply with context awareness.
Vui.COHOST is checkpoint with two speakers that can talk to each other.

# Voice Cloning

You can clone with the base model quite well but it's not perfect as hasn't seen that much audio / wasn't trained for long

# FAQ

1) Was developed with on two 4090's https://x.com/harrycblum/status/1752698806184063153
2) Hallucinations: yes the model does hallucinate, but this is the best I could do with limited resources! :(