Spaces:
Runtime error
Runtime error
samuelleecong
commited on
Commit
•
bda46a5
1
Parent(s):
1b46601
Update app.py
Browse files
app.py
CHANGED
@@ -16,7 +16,6 @@ device = "cuda:0" if torch.cuda.is_available() else "cpu"
|
|
16 |
|
17 |
asr_pipe = pipeline("automatic-speech-recognition", model="openai/whisper-base", device=device, generate_kwargs = {"task": "translate"})
|
18 |
|
19 |
-
|
20 |
m2m100_en_sw = pipeline('translation', 'facebook/m2m100_418M', src_lang='en', tgt_lang="sw")
|
21 |
|
22 |
# load text-to-speech checkpoint and speaker embeddings
|
@@ -55,8 +54,8 @@ def speech_to_speech_translation(audio):
|
|
55 |
|
56 |
title = "Cascaded STST"
|
57 |
description = """
|
58 |
-
Demo for cascaded speech-to-speech translation (STST), mapping from source speech in any language to target speech in
|
59 |
-
[SpeechT5 TTS](https://huggingface.co/
|
60 |
|
61 |
![Cascaded STST](https://huggingface.co/datasets/huggingface-course/audio-course-images/resolve/main/s2st_cascaded.png "Diagram of cascaded speech to speech translation")
|
62 |
"""
|
|
|
16 |
|
17 |
asr_pipe = pipeline("automatic-speech-recognition", model="openai/whisper-base", device=device, generate_kwargs = {"task": "translate"})
|
18 |
|
|
|
19 |
m2m100_en_sw = pipeline('translation', 'facebook/m2m100_418M', src_lang='en', tgt_lang="sw")
|
20 |
|
21 |
# load text-to-speech checkpoint and speaker embeddings
|
|
|
54 |
|
55 |
title = "Cascaded STST"
|
56 |
description = """
|
57 |
+
Demo for cascaded speech-to-speech translation (STST), mapping from source speech in any language to target speech in Swahili. Demo uses OpenAI's [Whisper Base](https://huggingface.co/openai/whisper-base) model for speech transcription, Meta's [M2M100](https://huggingface.co/facebook/m2m100_418M) for translation, and Microsoft's
|
58 |
+
[SpeechT5 TTS](https://huggingface.co/samuelleecong/speecht5_finetuned_swahili) that I fine-tuned model for Swahili text-to-speech:
|
59 |
|
60 |
![Cascaded STST](https://huggingface.co/datasets/huggingface-course/audio-course-images/resolve/main/s2st_cascaded.png "Diagram of cascaded speech to speech translation")
|
61 |
"""
|