Hugging Face's logo

Spaces:
vadhri
/
speech-to-speech-translation
Sleeping

App Files Community
vadhri commited on
Commit
b97b97e
•
1 Parent(s): 40d4679

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +4 -3
app.py CHANGED
@@ -46,9 +46,10 @@ def speech_to_speech_translation(audio):
46
 
47
  title = "Cascaded STST"
48
  description = """
49
- Demo for cascaded speech-to-speech translation (STST), mapping from source speech in any language to target speech in English.
50
- Demo uses OpenAI's [Whisper Base](https://huggingface.co/openai/whisper-base) model for speech translation, and a custom fine tuned version of Microsoft's
51
- [SpeechT5 TTS] fine tuned for NL - vadhri/speecht5_finetuned_voxpopuli_nl (https://huggingface.co/vadhri/speecht5_finetuned_voxpopuli_nl) model for text-to-speech:
 
52
 
53
  ![Cascaded STST](https://huggingface.co/datasets/huggingface-course/audio-course-images/resolve/main/s2st_cascaded.png "Diagram of cascaded speech to speech translation")
54
  """
 
46
 
47
  title = "Cascaded STST"
48
  description = """
49
+ Demo for cascaded speech-to-speech translation (STST), mapping from source speech in any language to target speech in Dutch.
50
+
51
+ Demo uses OpenAI's [Whisper Base](https://huggingface.co/openai/whisper-base) model for speech translation, and
52
+ a custom fine tuned version of [Microsoft's SpeechT5 TTS] fine tuned for Dutch - vadhri/speecht5_finetuned_voxpopuli_nl (https://huggingface.co/vadhri/speecht5_finetuned_voxpopuli_nl) model for text-to-speech:
53
 
54
  ![Cascaded STST](https://huggingface.co/datasets/huggingface-course/audio-course-images/resolve/main/s2st_cascaded.png "Diagram of cascaded speech to speech translation")
55
  """