Upload folder using huggingface_hub
Browse files- README.md +2 -2
- config.json +1 -1
- examples/m1.wav +0 -0
- examples/panchangjiang_gazi.wav +3 -0
README.md
CHANGED
|
@@ -44,8 +44,8 @@ import torchaudio
|
|
| 44 |
from transformers import AutoFeatureExtractor, AutoModel
|
| 45 |
|
| 46 |
# 1. Load the feature extractor and the codec model
|
| 47 |
-
feature_extractor = AutoFeatureExtractor.from_pretrained("
|
| 48 |
-
codec = AutoModel.from_pretrained("
|
| 49 |
|
| 50 |
# 2. Load and preprocess the audio
|
| 51 |
# The model expects a 16kHz sample rate.
|
|
|
|
| 44 |
from transformers import AutoFeatureExtractor, AutoModel
|
| 45 |
|
| 46 |
# 1. Load the feature extractor and the codec model
|
| 47 |
+
feature_extractor = AutoFeatureExtractor.from_pretrained("gaoyang07/XY_Tokenizer", trust_remote_code=True)
|
| 48 |
+
codec = AutoModel.from_pretrained("gaoyang07/XY_Tokenizer", trust_remote_code=True, device_map="auto").eval()
|
| 49 |
|
| 50 |
# 2. Load and preprocess the audio
|
| 51 |
# The model expects a 16kHz sample rate.
|
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"model_type": "
|
| 3 |
"input_sample_rate": 16000,
|
| 4 |
"output_sample_rate": 24000,
|
| 5 |
"encoder_downsample_rate": 1280,
|
|
|
|
| 1 |
{
|
| 2 |
+
"model_type": "xy_tokenizer",
|
| 3 |
"input_sample_rate": 16000,
|
| 4 |
"output_sample_rate": 24000,
|
| 5 |
"encoder_downsample_rate": 1280,
|
examples/m1.wav
CHANGED
|
Binary files a/examples/m1.wav and b/examples/m1.wav differ
|
|
|
examples/panchangjiang_gazi.wav
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ad00923005fa1c92a0cbd603c59aacb32828c9d6da73a331c08cd6756dd580a1
|
| 3 |
+
size 876844
|