Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -1,6 +1,8 @@
|
|
1 |
import torch
|
2 |
from transformers import pipeline
|
3 |
import json
|
|
|
|
|
4 |
|
5 |
transcribe = pipeline(task = "automatic-speech-recognition", model = "vasista22/whisper-tamil-medium", chunk_length_s=30, device="cpu")
|
6 |
transcribe.model.config.forced_decoder_ids = transcribe.tokenizer.get_decoder_prompt_ids(language="ta", task="transcribe")
|
@@ -31,6 +33,8 @@ def process_audio(params):
|
|
31 |
|
32 |
# text = response['results']['channels'][0]['alternatives'][0]['transcript']
|
33 |
print(audio)
|
|
|
|
|
34 |
text = transcribe(audio)["text"]
|
35 |
|
36 |
answer_dict = {}
|
|
|
1 |
import torch
|
2 |
from transformers import pipeline
|
3 |
import json
|
4 |
+
import requests
|
5 |
+
from io import BytesIO
|
6 |
|
7 |
transcribe = pipeline(task = "automatic-speech-recognition", model = "vasista22/whisper-tamil-medium", chunk_length_s=30, device="cpu")
|
8 |
transcribe.model.config.forced_decoder_ids = transcribe.tokenizer.get_decoder_prompt_ids(language="ta", task="transcribe")
|
|
|
33 |
|
34 |
# text = response['results']['channels'][0]['alternatives'][0]['transcript']
|
35 |
print(audio)
|
36 |
+
response = requests.get(audio_url)
|
37 |
+
audio = BytesIO(response.content)
|
38 |
text = transcribe(audio)["text"]
|
39 |
|
40 |
answer_dict = {}
|