Update README.md
Browse files
README.md
CHANGED
|
@@ -16,9 +16,28 @@ datasets:
|
|
| 16 |
- mythicinfinity/libritts_r
|
| 17 |
---
|
| 18 |
|
| 19 |
-
# NeuCodec
|
| 20 |
-
|
| 21 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
|
| 23 |
# Model Details
|
| 24 |
|
|
|
|
| 16 |
- mythicinfinity/libritts_r
|
| 17 |
---
|
| 18 |
|
| 19 |
+
# NeuCodec 🎧
|
| 20 |
+
|
| 21 |
+
[](https://www.youtube.com/watch?v=O7XH1lGZyYY)
|
| 22 |
+
<br>
|
| 23 |
+
*Click the image above to see NeuCodec in action on Youtube!*
|
| 24 |
+
|
| 25 |
+
*Created by Neuphonic - building faster, smaller, on-device voice AI*
|
| 26 |
+
|
| 27 |
+
A lightweight neural codec that encodes audio at just 0.8 kbps - perfect for researchers and builders who need something that *just works* for training high quality text-to-speech models.
|
| 28 |
+
|
| 29 |
+
# Key Features
|
| 30 |
+
|
| 31 |
+
🔊 Low bit-rate compression - a speech codec that compresses and reconstructs audio with near-inaudible reconstruction loss
|
| 32 |
+
<br>
|
| 33 |
+
🎼 Upsamples from 16kHz → 24kHz
|
| 34 |
+
<br>
|
| 35 |
+
🌍 Ready for real-world use - train your own SpeechLMs without needing to build your own codec
|
| 36 |
+
<br>
|
| 37 |
+
🏢 Commercial use permitted - use it in your own tools or products
|
| 38 |
+
<br>
|
| 39 |
+
📊 Released with large pre-encoded datasets - we’ve compressed Emilia-YODAS from 1.7TB to 41GB using NeuCodec, significantly reducing the compute requirements needed for training
|
| 40 |
+
<br>
|
| 41 |
|
| 42 |
# Model Details
|
| 43 |
|