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.
State-of-the-art deep learning based audio codec supporting both mono 24 kHz audio and stereo 48 kHz audio.
| Date | Stars |
|---|---|
| 2026-07-31 | 4006 |
| 2026-08-06 | 4011 |
| 2026-08-08 | 4012 |
| 2026-08-10 | 4013 |
| 2026-08-11 | 4013 |
| 2026-08-12 | 4013 |
| 2026-08-13 | 4013 |
| 2026-08-14 | 4014 |
| 2026-08-15 | 4015 |
| 2026-08-16 | 4015 |
| 2026-08-17 | 4016 |
| 2026-08-18 | 4023 |
| 2026-08-19 | 4029 |
| 2026-08-20 | 4031 |
| 2026-08-21 | 4034 |
| 2026-08-23 | 4035 |
| 2026-08-24 | 4037 |
| 2026-08-25 | 4040 |
| 2026-08-27 | 4041 |
| 2026-08-28 | 4041 |
| 2026-08-29 | 4042 |
| 2026-08-31 | 4044 |
| 2026-09-01 | 4045 |
| 2026-09-03 | 4046 |
| 2026-09-04 | 4047 |
| 2026-09-05 | 4050 |
| 2026-09-06 | 4051 |
| 2026-09-07 | 4052 |
| 2026-09-08 | 4051 |
| 2026-09-09 | 4053 |
| 2026-09-10 | 4054 |
| 2026-09-11 | 4055 |
| 2026-09-13 | 4056 |
| 2026-09-16 | 4058 |
| 2026-09-17 | 4058 |
| 2026-09-18 | 4056 |
| 2026-09-20 | 4056 |
| 2026-09-21 | 4057 |
| 2026-09-23 | 4058 |
| 2026-09-24 | 4059 |
| 2026-09-25 | 4060 |
| 2026-09-28 | 4059 |
| 2026-09-30 | 4059 |
| 2026-10-01 | 4060 |
| 2026-10-02 | 4060 |
| 2026-10-05 | 4059 |
| 2026-10-06 | 4062 |
| 2026-10-12 | 4063 |
| 2026-10-14 | 4065 |
| 2026-10-16 | 4066 |
| 2026-10-21 | 4068 |
| 2026-10-22 | 4066 |
| 2026-10-24 | 4067 |
| 2026-10-25 | 4069 |
| 2026-10-28 | 4070 |
| 2026-10-29 | 4072 |
| 2026-10-31 | 4073 |
| 2026-11-01 | 4073 |
| 2026-11-03 | 4075 |
| 2026-11-04 | 4075 |
Today
— stars today
This week
+5 stars this week
This month
+16 stars this month
Momentum
0.0
growth rate 0.12%/day
# EnCodec: High Fidelity Neural Audio Compression
This is the code for the EnCodec neural codec presented in the [High Fidelity Neural Audio Compression](https://arxiv.org/pdf/2210.13438.pdf) [[abs]](https://arxiv.org/abs/2210.13438).
paper. We provide our two multi-bandwidth models:
* A causal model operating at 24 kHz on monophonic audio trained on a variety of audio data.
* A non-causal model operating at 48 kHz on stereophonic audio trained on music-only data.
The 24 kHz model can compress to 1.5, 3, 6, 12 or 24 kbps, while the 48 kHz model
support 3, 6, 12 and 24 kbps. We also provide a pre-trained language model for each
of the models, that can further compress the representation by up to 40% without
any further loss of quality.
For reference, we also provide the code for our novel [MS-STFT discriminator](encodec/msstftd.py) and the [balancer](encodec/balancer.py).
<p align="center">
<img src="./architecture.png" alt="Schema representing the structure of Encodec,
with a convolutional+LSTM encoder, a Residual Vector Quantization in the middle,
followed by a convolutional+LSTM decoder. A multiscale complex spectrogram discriminator is applied to the output, along with objective reconstruction losses.
A small transformer model is trained to predict the RVQ output."
width="800px"></p>
## Samples
Samples including baselines are provided on [our sample page](https://ai.honu.io/papers/encodec/samples.html).
You can also have a quick demo of what we achieve for 48 kHz music with EnCodec, along with
entropy coding, by clicking the thumbnail (original tracks provided by [Lucille Crew](https://open.spotify.com/artist/5eLv7rNfrf3IjMnK311ByP?si=X_zD9ackRRGjFP5Y6Q7Zng) and [Voyageur I](https://open.spotify.com/artist/21HymveeIhDcM4KDKeNLz0?si=4zXF8VpeQpeKR9QUIuck9Q)).
<p align="center">
<a href="https://ai.honu.io/papers/encodec/final.mp4">
<img src="./thumbnail.png" alt="Thumbnail for the sample video.
You will first here the ground truth, then ~3kbps, then 12kbps, for two songs."></a></p>
## 🤗 Transformers
Encodec has now been added to Transformers. For more information, please refer to [Transformers' Encodec docs](https://huggingface.co/docs/transformers/main/en/model_doc/encodec).
You can find both the [24KHz](https://huggingface.co/facebook/encodec_24khz) and [48KHz](https://huggingface.co/facebook/encodec_48khz) checkpoints on the 🤗 Hub.
Using 🤗 Transformers, you can leverage Encodec at scale along with all the other supported models and datasets. ⚡️
Alternatively you can also directly use the encodec package, as detailed in the Usage section.
To use first you'd need to set up your development environment!
```
pip install -U datasets
pip install git+https://github.com/huggingface/transformers.git@main
```
Then, start embedding your audio datasets at scale!
```python
from datasets import load_dataset, Audio
from transformers import EncodecModel, AutoProcessor
# dummy dataset, however you can swap this with an dataset on the 🤗 hub or bring your own
librispeech_dummy = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation")
# load the model + processor (for pre-processing the audio)
model = EncodecModel.from_pretrained("facebook/encodec_24khz")
processor = AutoProcessor.from_pretrained("facebook/encodec_24khz")
# cast the audio data to the correct sampling rate for the model
librispeech_dummy = librispeech_dummy.cast_column("audio", Audio(sampling_rate=processor.sampling_rate))
audio_sample = librispeech_dummy[0]["audio"]["array"]
# pre-process the inputs
inputs = processor(raw_audio=audio_sample, sampling_rate=processor.sampling_rate, return_tensors="pt")
# explicitly encode then decode the audio inputs
encoder_outputs = model.encode(inputs["input_values"], inputs["padding_mask"])
audio_valueExcerpt of 10,158 characters
Read on GitHub
Alexandre Défossez · France
24
Ikko Eltociear Ashimine · Japan
1
1
vb · @openai
1
1
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:cb44ba4e7cf9c8fd, llm:Repository description: 'State-of-the-art deep learning based audio codec supporting both mono 24 kHz audio and stereo 48 kHz audio.'
matched fp:cb44ba4e7cf9c8fd, llm:Repository description: 'State-of-the-art deep learning based audio codec supporting both mono 24 kHz audio and stereo 48 kHz audio.'