Link the community audio.cpp GGUF builds

#3
by sw005320 - opened
Files changed (1) hide show
  1. README.md +2 -0
README.md CHANGED
@@ -39,6 +39,8 @@ The code, pre-trained model weights, and training logs are publicly released to
39
  Inference examples can be found on our [project page](https://www.wavlab.org/activities/2024/owsm/).
40
  The Gradio demo is [here](https://huggingface.co/spaces/espnet/owsm-v4).
41
 
 
 
42
  [OWSM v4](https://arxiv.org/abs/2506.00338) is the latest version in the OWSM series, which significantly outperforms OWSM v3.1 in LID and multilingual ASR.
43
  Additionally, OWSM v4 applies 8 times subsampling (instead of 4 times in OWSM v3.1) to the log Mel features, leading to a final resolution of 80 ms in the encoder.
44
  When running inference, we recommend setting `maxlenratio=1.0` (default) instead of smaller values.
 
39
  Inference examples can be found on our [project page](https://www.wavlab.org/activities/2024/owsm/).
40
  The Gradio demo is [here](https://huggingface.co/spaces/espnet/owsm-v4).
41
 
42
+ For CPU inference without Python, GGUF conversions of this model are available at [audio-cpp/audio.cpp-gguf](https://huggingface.co/audio-cpp/audio.cpp-gguf/tree/main/OWSM-GGUF) (`owsm-v4-medium-1b`), for use with [audio.cpp](https://github.com/0xShug0/audio.cpp). These builds are produced and maintained outside ESPnet.
43
+
44
  [OWSM v4](https://arxiv.org/abs/2506.00338) is the latest version in the OWSM series, which significantly outperforms OWSM v3.1 in LID and multilingual ASR.
45
  Additionally, OWSM v4 applies 8 times subsampling (instead of 4 times in OWSM v3.1) to the log Mel features, leading to a final resolution of 80 ms in the encoder.
46
  When running inference, we recommend setting `maxlenratio=1.0` (default) instead of smaller values.