Upload folder using huggingface_hub

README.md

@@ -0,0 +1,14 @@
+---
+library_name: mlx-audio-plus
+base_model:
+- FunAudioLLM/Fun-CosyVoice3-0.5B-2512
+tags:
+- mlx
+- speech-tokenizer
+---
+
+# mlx-community/S3TokenizerV3
+
+S3TokenizerV3 (Supervised Semantic Speech Tokenizer) converted to MLX format from [FunAudioLLM/Fun-CosyVoice3-0.5B-2512](https://huggingface.co/FunAudioLLM/Fun-CosyVoice3-0.5B-2512).
+
+This tokenizer is automatically downloaded when using CosyVoice 3 with [mlx-audio-plus](https://github.com/DePasqualeOrg/mlx-audio-plus) version **0.1.4**.

config.json

@@ -0,0 +1,9 @@
+{
+  "model_type": "s3_tokenizer_v3",
+  "n_mels": 128,
+  "n_audio_ctx": 1500,
+  "n_audio_state": 1280,
+  "n_audio_head": 20,
+  "n_audio_layer": 12,
+  "n_codebook_size": 6561
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee05a65d4472f089d55b5dfcd9e19f42e3245f4f19d612b90c04cb051a80105e
+size 968059321