facebook
/

xm_transformer_s2ut_800m-es-en-st-asr-bt_h1_2022

Audio-to-Audio Fairseq

audio speech-to-speech-translation

Model card Files Files and versions Community

lpw commited on Sep 30, 2022

Commit

1913999

•

1 Parent(s): 3c9445c

Update config.yaml

Files changed (1) hide show

config.yaml +3 -1

config.yaml CHANGED Viewed

@@ -8,4 +8,6 @@ hub:
   input_type: standardized_waveform
   unit_vocoder: true
   tts_model_id: pytorch/fairseq:unit_hifigan_mhubert_vp_en_es_fr_it3_400k_layer11_km1000_lj_dur

   input_type: standardized_waveform
   unit_vocoder: true
   tts_model_id: pytorch/fairseq:unit_hifigan_mhubert_vp_en_es_fr_it3_400k_layer11_km1000_lj_dur
+  generation_args:
+    beam: 1
+    max_len_a: 0.03125