Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
A Fast TTS Engine
| Date | Stars |
|---|---|
| 2026-07-30 | 627 |
| 2026-08-07 | 626 |
| 2026-08-18 | 625 |
| 2026-08-27 | 625 |
| 2026-09-17 | 624 |
| 2026-09-21 | 623 |
| 2026-09-29 | 622 |
| 2026-10-05 | 621 |
| 2026-10-17 | 622 |
| 2026-10-18 | 623 |
| 2026-10-24 | 622 |
| 2026-10-27 | 623 |
| 2026-10-28 | 623 |
| 2026-11-04 | 623 |
Today
— stars today
This week
— stars this week
This month
+2 stars this month
Momentum
0.0
growth rate 0.00%/day
[](https://discord.gg/https://discord.gg/BEMVTmcPEs)
# Auralis 🌌 (/auˈralis/)
Transform text into natural speech (with voice cloning) at warp speed. Process an entire novel in minutes, not hours.
## What is Auralis? 🚀
Auralis is a text-to-speech engine that makes voice generation practical for real-world use:
- Convert the entire first Harry Potter book to speech in 10 minutes (**realtime factor of ≈ 0.02x!** )
- Automatically enhance the reference quality, you can register them even with a low quality mic!
- It can be configured to have a small memory footprint (scheduler_max_concurrency)
- Process multiple requests simultaneously
- Stream long texts piece by piece
## Quick Start ⭐
1. Create a new Conda environment:
```bash
conda create -n auralis_env python=3.10 -y
```
2. Activate the environment:
```bash
conda activate auralis_env
```
3. Install Auralis:
```bash
pip install auralis
```
and then you can try it out via **python**
```python
from auralis import TTS, TTSRequest
# Initialize
tts = TTS().from_pretrained("AstraMindAI/xttsv2", gpt_model='AstraMindAI/xtts2-gpt')
# Generate speech
request = TTSRequest(
text="Hello Earth! This is Auralis speaking.",
speaker_files=['reference.wav']
)
output = tts.generate_speech(request)
output.save('hello.wav')
```
or via **cli** using the openai compatible server
```commandline
auralis.openai --host 127.0.0.1 --port 8000 --model AstraMindAI/xttsv2 --gpt_model AstraMindAI/xtts2-gpt --max_concurrency 8 --vllm_logging_level warn
```
You can see [here](https://github.com/astramind-ai/Auralis/tree/main/docs/USING_OAI_SERVER.md) for a more in-depth explanation or try it out with this [example](https://github.com/astramind-ai/Auralis/tree/main/examples/use_openai_server.py)
## Key Features 🛸
### Speed & Efficiency
- Processes long texts rapidly using smart batching
- Runs on consumer GPUs without memory issues
- Handles multiple requests in parallel
### Easy Integration
- Simple Python API
- Streaming support for long texts
- Built-in audio enhancement
- Automatic language detection
### Audio Quality
- Voice cloning from short samples
- Background noise reduction
- Speech clarity enhancement
- Volume normalization
## XTTSv2 Finetunes
You can use your own XTTSv2 finetunes by simply converting them from the standard coqui checkpoint format to our safetensor format. Use [this script](https://github.com/astramind-ai/Auralis/blob/main/src/auralis/models/xttsv2/utils/checkpoint_converter.py):
```commandline
python checkpoint_converter.py path/to/checkpoint.pth --output_dir path/to/output
```
it will create two folders, one with the core xttsv2 checkpoint and one with the gtp2 component. Then create a TTS instance with
```python
tts = TTS().from_pretrained("som/core-xttsv2_model", gpt_model='some/xttsv2-gpt_model')
```
## Examples & Usage 🚀
### Basic Examples ⭐
<details>
<summary><b>Simple Text Generation</b></summary>
```python
from auralis import TTS, TTSRequest
# Initialize
tts = TTS().from_pretrained("AstraMindAI/xttsv2", gpt_model='AstraMindAI/xtts2-gpt')
# Basic generation
request = TTSRequest(
text="Hello Earth! This is Auralis speaking.",
speaker_files=["speaker.wav"]
)
output = tts.generate_speech(request)
output.save("hello.wav")
```
</details>
<details>
<summary><b>Working with TTSRequest</b> 🎤</summary>
```python
# Basic request
request = TTSRequest(
text="Hello world!",
speaker_files=["speaker.wav"]
)
# Enhanced audio processing
request = TTSRequest(
text="Pristine audio quality",
speaker_files=["speaker.wav"],
audio_config=AudioPreprocessingConfig(
normalize=True,
trim_silence=True,
enhance_speech=True,
enhance_amount=1.5
)
)
# Language-specific request
request = TTSRequest(
text="Bonjour le monde!",
speaker_files=["speaker.wav"],
language="fr"
)
# Streaming configuExcerpt of 13,338 characters
Read on GitHub
118
16
藍+85CD · @moeru-ai
8
1
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:7346bda1cbd7e2eb, topic:tts, readme:text-to-speech, readme:voice cloning