Update README.md

README.md

@@ -99,7 +99,7 @@ The following hyperparameters were used during training:
 - Datasets 1.17.1.dev0
 - Tokenizers 0.10.3
 
-### Inference Without Decoder
+### Inference With LM
 
 ```python
 import torch
@@ -108,9 +108,11 @@ from transformers import AutoModelForCTC, AutoProcessor
 import torchaudio.functional as F
 
 
-model_id = "patrickvonplaten/wav2vec2-large-xlsr-53-spanish-with-lm"
+model_id = "hf-test/xls-r-300m-sv"
 
-sample = next(iter(load_dataset("common_voice", "es", split="test", streaming=True)))
+sample_iter = iter(load_dataset("mozilla-foundation/common_voice_7_0", "sv-SE", split="test", streaming=True, use_auth_token=True))
+
+sample = next(sample_iter)
 resampled_audio = F.resample(torch.tensor(sample["audio"]["array"]), 48_000, 16_000).numpy()
 
 model = AutoModelForCTC.from_pretrained(model_id)
@@ -121,15 +123,9 @@ input_values = processor(resampled_audio, return_tensors="pt").input_values
 with torch.no_grad():
     logits = model(input_values).logits
 
--prediction_ids = torch.argmax(logits, dim=-1)
--transcription = processor.batch_decode(prediction_ids)
-+transcription = processor.batch_decode(logits.numpy()).text
+transcription = processor.batch_decode(logits.numpy()).text
 ```
 
-
-### Inference With Decoder
-
-
 ### Eval results on Common Voice 7 "test":
 
 **Without LM**: 27.30 WER