---
license: cc-by-nc-sa-4.0
base_model:
- fishaudio/openaudio-s1-mini
pipeline_tag: text-to-speech
tags:
- gguf-connector
---
## gguf quantized version of openaudio
- base model from [fishaudio](https://huggingface.co/fishaudio)
- text-to-speech synthesis

### **run it with gguf-connector**
```
ggc o2
```

![screenshot](https://raw.githubusercontent.com/calcuis/text-to-speech-synthesis-2/master/demo.png)

| Prompt | Audio Sample |
|--------|---------------|
|`Hey Connector, why your appearance looks so stupid?`<br/>`Oh, really? maybe I ate too much smart beans.`<br/>`Wow. Amazing (laughing).`<br/>`Let's go to get some more smart beans and you will become stupid as well.`<br/> | 🎧 **audio-sample-1**<br><audio controls src="https://huggingface.co/calcuis/openaudio-gguf/resolve/main/samples%5Caudio1.wav"></audio> |
|`Suddenly the plane's engines began failing, and the pilot says there isn't much time, and he'll keep the plane in the air as long as he can, and told his two passengers to take the only two parachutes on board and bail out. The world's smartest man immediately took a parachute and said "I'm the world's smartest man! The world needs me, so I can't die here!", and then jumped out of the plane. The pilot tells the hippie to hurry up and take the other parachute, because there aren't any more. And the hippie says "Relax man. We'll be fine. The world's smartest man took my backpack."`<br/> | 🎧 **audio-sample-2**<br><audio controls src="https://huggingface.co/calcuis/openaudio-gguf/resolve/main/samples%5Caudio2.wav"></audio> |

### **review/reference**
- simply execute the command (`ggc o2`) above in console/terminal
- opt a `codec` and a `model` gguf in the current directory to interact with (see example below)

>
>GGUF file(s) available. Select which one for **codec**:
>
>1. codec-q2_k.gguf
>2. codec-q3_k_m.gguf
>3. codec-q4_k_m.gguf (recommended)
>4. codec-q5_k_m.gguf
>5. codec-q6_k.gguf
>6. model-bf16.gguf
>7. model-f16.gguf
>8. model-f32.gguf
>
>Enter your choice (1 to 8): 3
> 
>GGUF file(s) available. Select which one for **model**:
>
>1. codec-q2_k.gguf
>2. codec-q3_k_m.gguf
>3. codec-q4_k_m.gguf
>4. codec-q5_k_m.gguf
>5. codec-q6_k.gguf
>6. model-bf16.gguf (recommended)
>7. model-f16.gguf (for non-cuda user)
>8. model-f32.gguf
>
>Enter your choice (1 to 8): _
>

- note: for the latest update, only tokenizer will be pulled to `models/fish` automatically during the first launch, you need to prepare the **codec** and **model** files yourself, working like [vision](https://huggingface.co/calcuis/llava-gguf) connector right away; mix and match, more flexible
- run it entirely offline; i.e., from local URL:  http://127.0.0.1:7860 with lazy webui
- gguf-connector ([pypi](https://pypi.org/project/gguf-connector))