Update README.md
Browse files
README.md
CHANGED
|
@@ -8,18 +8,17 @@ tags:
|
|
| 8 |
- tokenizer
|
| 9 |
- codec-representation
|
| 10 |
---
|
| 11 |
-
# WavTokenizer
|
| 12 |
-
SOTA Discrete Codec Models With Forty Tokens Per Second for Audio Language Modeling
|
| 13 |
|
| 14 |
|
| 15 |
|
| 16 |
[](https://github.com/jishengpeng/wavtokenizer)
|
| 17 |
[](https://wavtokenizer.github.io/)
|
| 18 |
-
[](https://
|
| 19 |
|
| 20 |
|
| 21 |
|
| 22 |
-
### ππ with WavTokenizer, you can represent speech, music, and audio with only 40 tokens
|
| 23 |
### ππ with WavTokenizer, You can get strong reconstruction results.
|
| 24 |
### ππ WavTokenizer owns rich semantic information and is build for audio language models such as GPT4-o.
|
| 25 |
|
|
@@ -110,8 +109,8 @@ audio_out = wavtokenizer.decode(features, bandwidth_id=bandwidth_id)
|
|
| 110 |
|
| 111 |
| Model name | HuggingFace | Corpus | aa | Parameters | Open-Source |
|
| 112 |
|:--------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------:|:--------:|:---------:|:----------:|:------:|
|
| 113 |
-
| WavTokenizer-small-600-24k-4096 | [π€](https://
|
| 114 |
-
| WavTokenizer-small-320-24k-4096 | [π€](https://
|
| 115 |
| WavTokenizer-medium-600-24k-4096 | [π€](https://github.com/jishengpeng/wavtokenizer) | 10000 Hours | 40 | Speech, Audio, Music | Coming Soon|
|
| 116 |
| WavTokenizer-medium-320-24k-4096 | [π€](https://github.com/jishengpeng/wavtokenizer) | 10000 Hours | 75 | Speech, Audio, Music | Coming Soon|
|
| 117 |
| WavTokenizer-large-600-24k-4096 | [π€](https://github.com/jishengpeng/wavtokenizer) | LibriTTS | 40 | Speech, Audio, Music | Coming Soon|
|
|
|
|
| 8 |
- tokenizer
|
| 9 |
- codec-representation
|
| 10 |
---
|
| 11 |
+
# WavTokenizer: SOTA Discrete Codec Models With Forty Tokens Per Second for Audio Language Modeling
|
|
|
|
| 12 |
|
| 13 |
|
| 14 |
|
| 15 |
[](https://github.com/jishengpeng/wavtokenizer)
|
| 16 |
[](https://wavtokenizer.github.io/)
|
| 17 |
+
[](https://huggingface.co/novateur/WavTokenizer)
|
| 18 |
|
| 19 |
|
| 20 |
|
| 21 |
+
### ππ with WavTokenizer, you can represent speech, music, and audio with only 40 tokens per second!
|
| 22 |
### ππ with WavTokenizer, You can get strong reconstruction results.
|
| 23 |
### ππ WavTokenizer owns rich semantic information and is build for audio language models such as GPT4-o.
|
| 24 |
|
|
|
|
| 109 |
|
| 110 |
| Model name | HuggingFace | Corpus | aa | Parameters | Open-Source |
|
| 111 |
|:--------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------:|:--------:|:---------:|:----------:|:------:|
|
| 112 |
+
| WavTokenizer-small-600-24k-4096 | [π€](https://huggingface.co/novateur/WavTokenizer/blob/main/WavTokenizer_small_600_24k_4096.ckpt) | LibriTTS | 40 | Speech | β |
|
| 113 |
+
| WavTokenizer-small-320-24k-4096 | [π€](https://huggingface.co/novateur/WavTokenizer/blob/main/WavTokenizer_small_320_24k_4096.ckpt) | LibriTTS | 75 | Speech | β|
|
| 114 |
| WavTokenizer-medium-600-24k-4096 | [π€](https://github.com/jishengpeng/wavtokenizer) | 10000 Hours | 40 | Speech, Audio, Music | Coming Soon|
|
| 115 |
| WavTokenizer-medium-320-24k-4096 | [π€](https://github.com/jishengpeng/wavtokenizer) | 10000 Hours | 75 | Speech, Audio, Music | Coming Soon|
|
| 116 |
| WavTokenizer-large-600-24k-4096 | [π€](https://github.com/jishengpeng/wavtokenizer) | LibriTTS | 40 | Speech, Audio, Music | Coming Soon|
|