Hugging Face's logo

Spaces:
rudyong
/
speech-to-speech-translation_en-nl
Runtime error

App Files Community
Rudy Ong commited on
Commit
9c5dc07
·
1 Parent(s): 00df10b

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +1 -2
app.py CHANGED
@@ -13,7 +13,6 @@ asr_pipe = pipeline("automatic-speech-recognition", model="openai/whisper-base",
13
 
14
  # load text-to-speech checkpoint and speaker embeddings
15
  processor = SpeechT5Processor.from_pretrained("microsoft/speecht5_tts")
16
-
17
  model = SpeechT5ForTextToSpeech.from_pretrained("Bolakubus/speecht5_finetuned_voxpopuli_nl").to(device)
18
  vocoder = SpeechT5HifiGan.from_pretrained("microsoft/speecht5_hifigan").to(device)
19
 
@@ -41,7 +40,7 @@ def speech_to_speech_translation(audio):
41
 
42
  title = "Cascaded STST"
43
  description = """
44
- Demo for cascaded speech-to-speech translation (STST), mapping from source speech in any language to target speech in English. Demo uses OpenAI's [Whisper Base](https://huggingface.co/openai/whisper-base) model for speech translation, and Microsoft's
45
  [SpeechT5 TTS](https://huggingface.co/microsoft/speecht5_tts) model for text-to-speech:
46
 
47
  ![Cascaded STST](https://huggingface.co/datasets/huggingface-course/audio-course-images/resolve/main/s2st_cascaded.png "Diagram of cascaded speech to speech translation")
 
13
 
14
  # load text-to-speech checkpoint and speaker embeddings
15
  processor = SpeechT5Processor.from_pretrained("microsoft/speecht5_tts")
 
16
  model = SpeechT5ForTextToSpeech.from_pretrained("Bolakubus/speecht5_finetuned_voxpopuli_nl").to(device)
17
  vocoder = SpeechT5HifiGan.from_pretrained("microsoft/speecht5_hifigan").to(device)
18
 
 
40
 
41
  title = "Cascaded STST"
42
  description = """
43
+ Demo for cascaded speech-to-speech translation (STST), mapping from source speech in any language to target speech in Dutch. Demo uses OpenAI's [Whisper Base](https://huggingface.co/openai/whisper-base) model for speech translation, and Microsoft's
44
  [SpeechT5 TTS](https://huggingface.co/microsoft/speecht5_tts) model for text-to-speech:
45
 
46
  ![Cascaded STST](https://huggingface.co/datasets/huggingface-course/audio-course-images/resolve/main/s2st_cascaded.png "Diagram of cascaded speech to speech translation")