Alexndem commited on
Commit
9bf9f5e
1 Parent(s): 4127c5c

Upload app.py

Browse files
Files changed (1) hide show
  1. app.py +4 -2
app.py CHANGED
@@ -70,8 +70,10 @@ def speech_to_speech_translation(audio):
70
 
71
  title = "Cascaded STST"
72
  description = """
73
- 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
74
- [SpeechT5 TTS](https://huggingface.co/microsoft/speecht5_tts) model for text-to-speech:
 
 
75
  ![Cascaded STST](https://huggingface.co/datasets/huggingface-course/audio-course-images/resolve/main/s2st_cascaded.png "Diagram of cascaded speech to speech translation")
76
  """
77
 
 
70
 
71
  title = "Cascaded STST"
72
  description = """
73
+ * Сначала модель распознает речь с помощью voidful/wav2vec2-xlsr-multilingual-56 и возвращает текст на любом из 56 языков.
74
+ * Далее происходит перевод текста с любого на английский с помощью Helsinki-NLP/opus-mt-mul-en, а затем с английского на русский также с помощью Helsinki-NLP/opus-mt-en-ru
75
+ * В конце осуществляется воспроизведение русского текста моделью facebook/mms-tts-rus
76
+ Demo for cascaded speech-to-speech translation (STST), mapping from source speech in any language to target speech in Russian. Demo uses facebook/mms-tts-rus model for text-to-speech:
77
  ![Cascaded STST](https://huggingface.co/datasets/huggingface-course/audio-course-images/resolve/main/s2st_cascaded.png "Diagram of cascaded speech to speech translation")
78
  """
79