Update README.md

README.md

@@ -13,11 +13,9 @@ datasets:
 - covost2
 - europarl_st
 - voxpopuli
-widget:
-- example_title: Common Voice sample 1
-  src: https://huggingface.co/facebook/xm_transformer_600m-es_en-multi_domain/resolve/main/common_voice_es_19966634.flac
+
 ---
-## xm_transformer_s2ut_800m-es-en-st-asr-bt_h1_2022
+## xm_transformer_unity_en-hk
 
 Speech-to-speech translation model from fairseq S2UT ([paper](https://arxiv.org/abs/2204.02967)/[code](https://github.com/facebookresearch/fairseq/blob/main/examples/speech_to_speech/docs/enhanced_direct_s2st_discrete_units.md)):
 - Spanish-English
@@ -43,7 +41,7 @@ import torchaudio
 cache_dir = os.getenv("HUGGINGFACE_HUB_CACHE")
 
 models, cfg, task = load_model_ensemble_and_task_from_hf_hub(
-    "facebook/xm_transformer_s2ut_800m-es-en-st-asr-bt_h1_2022",
+    "facebook/xm_transformer_unity_en-hk",
     arg_overrides={"config_yaml": "config.yaml", "task": "speech_to_text"},
     cache_dir=cache_dir,
 )
@@ -64,7 +62,7 @@ cache_dir = (
     cache_dir or (Path.home() / ".cache" / library_name).as_posix()
 )
 cache_dir = snapshot_download(
-    f"facebook/unit_hifigan_mhubert_vp_en_es_fr_it3_400k_layer11_km1000_lj_dur", cache_dir=cache_dir, library_name=library_name
+    f"facebook/unit_hifigan_HK_layer12.km2500_frame_TAT-TTS", cache_dir=cache_dir, library_name=library_name
 )
 
 x = hub_utils.from_pretrained(