Marco-Cheung commited on
Commit
71c7a94
1 Parent(s): de411e0

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +6 -1
app.py CHANGED
@@ -1,7 +1,7 @@
1
  import gradio as gr
2
  import numpy as np
3
  import torch
4
- from transformers import AutoProcessor, pipeline, BarkModel
5
 
6
  ASR_MODEL_NAME = "bofenghuang/whisper-large-v2-cv11-german"
7
  TTS_MODEL_NAME = "suno/bark-small"
@@ -18,6 +18,11 @@ asr_pipe = pipeline("automatic-speech-recognition", model=ASR_MODEL_NAME, chunk_
18
  # set the forced ids
19
  asr_pipe.model.config.forced_decoder_ids = asr_pipe.tokenizer.get_decoder_prompt_ids(language='de', task="translate")
20
 
 
 
 
 
 
21
  # load text-to-speech checkpoint
22
  processor = AutoProcessor.from_pretrained("suno/bark-small")
23
  model = BarkModel.from_pretrained("suno/bark-small").to(device)
 
1
  import gradio as gr
2
  import numpy as np
3
  import torch
4
+ from transformers import AutoProcessor, pipeline, BarkModel, GenerationConfig
5
 
6
  ASR_MODEL_NAME = "bofenghuang/whisper-large-v2-cv11-german"
7
  TTS_MODEL_NAME = "suno/bark-small"
 
18
  # set the forced ids
19
  asr_pipe.model.config.forced_decoder_ids = asr_pipe.tokenizer.get_decoder_prompt_ids(language='de', task="translate")
20
 
21
+ MULTILINGUAL = True # set True for multilingual models, False for English-only
22
+
23
+ if MULTILINGUAL:
24
+ generation_config = GenerationConfig.from_pretrained("openai/whisper-large-v2")
25
+
26
  # load text-to-speech checkpoint
27
  processor = AutoProcessor.from_pretrained("suno/bark-small")
28
  model = BarkModel.from_pretrained("suno/bark-small").to(device)