Skip to content

Commit

Permalink
UPDATES
Browse files Browse the repository at this point in the history
  • Loading branch information
Anantha Raju C committed Feb 25, 2024
1 parent 816ccaa commit a63d25a
Show file tree
Hide file tree
Showing 4 changed files with 129 additions and 29 deletions.
61 changes: 42 additions & 19 deletions codePython/nlp1.py
View file
Original file line number Diff line number Diff line change
@@ -1,31 +1,53 @@
import torch
from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline
from datasets import load_dataset
import speech_recognition as sr

import numpy as np
import sys
import json

# Access command-line arguments
arguments = sys.argv[1:]

# Print the arguments passed from Java
# Create an empty dictionary to store key-value pairs
args_dict = {}

# Parse arguments with keys
for arg in arguments:
print("Argument:", arg)
key, value = arg.split('=') # Split the argument at '=' to separate key and value
args_dict[key] = value

# Define the key of interest
audio_file_path = 'audio_file_path'

# Check if the desired key exists in the dictionary
if audio_file_path in args_dict:
audio_file_path_value = args_dict[audio_file_path]
print(f"Value for key '{audio_file_path}': {audio_file_path_value}")
else:
print(f"Key '{audio_file_path}' not found in the arguments.")

# device = "cuda:0" if torch.cuda.is_available() else "cpu"
device = "cpu"
# Check device availability
device = "cuda:0" if torch.cuda.is_available() else "cpu"

# Determine the torch data type based on device availability
torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32

# Define the model ID - https://huggingface.co/openai/whisper-large-v3
model_id = "openai/whisper-large-v3"

# Load the speech-to-text model
model = AutoModelForSpeechSeq2Seq.from_pretrained(
model_id, torch_dtype=torch_dtype, low_cpu_mem_usage=True, use_safetensors=True
model_id,
torch_dtype=torch_dtype,
low_cpu_mem_usage=True,
use_safetensors=True
)
model.to(device)

# Load the processor
processor = AutoProcessor.from_pretrained(model_id)

# Create a pipeline for automatic speech recognition
pipe = pipeline(
"automatic-speech-recognition",
model=model,
Expand All @@ -39,21 +61,22 @@
device=device,
)

# Perform automatic speech recognition
resultTranscription = pipe(audio_file_path_value, generate_kwargs={"language": None})
print("\n",resultTranscription["text"])

resultTranscription = pipe("/home/anantha/PCF/AIML/SpeechToText/HIN_M_AbhishekS.mp3", generate_kwargs={"language": None})
print("")
print(resultTranscription["text"])


resultTranslation = pipe("/home/anantha/PCF/AIML/SpeechToText/HIN_M_AbhishekS.mp3", generate_kwargs={"task": "translate"})
print("")
print(resultTranslation["text"])
# Perform translation
resultTranslation = pipe(audio_file_path_value, generate_kwargs={"task": "translate"})
print("\n",resultTranslation["text"])

# Load the sentiment analysis classifier
classifier = pipeline("sentiment-analysis",
model="distilbert/distilbert-base-uncased-finetuned-sst-2-english",
revision ="af0f99b")

classifier = pipeline("sentiment-analysis",model="distilbert/distilbert-base-uncased-finetuned-sst-2-english",revision ="af0f99b")
# Perform sentiment analysis on the translated text
resultClassifier = classifier(resultTranslation["text"])
print("")
print(resultClassifier)
print("\n",resultClassifier)

# Create a Python dictionary
data = {
Expand All @@ -66,4 +89,4 @@
json_string = json.dumps(data)

# Print the JSON string
print(json_string)
print("\n",json_string)
60 changes: 60 additions & 0 deletions codePython/program_output.json
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,60 @@
{
"resultTranscription": {
"text": " इसके बाद हम एक ऐसे खिलाड़ी के बारे में बात करेंगे जिससे इस वर्ल्ड कप में भारतिय फैंस को काफी उमीदे रहेंगे इस खिलाड़ी के बारे में मैं आपको बस इतना बता दूं कि ये वो खिलाड़ी हैं जिसने इंटरनेशनल क्रिकेट में अपनी एंट्री का एलान औस्ट्रेलिया के गेंज जोरदार प्रदर्शन से किया आज से वर्ल्ड कप के दारान हर रोज हम आपको मिलते रहेंगे EAM Cricket World Cup 2007 अपडेट में",
"chunks": [
{
"timestamp": [
0.0,
5.42
],
"text": " इसके बाद हम एक ऐसे खिलाड़ी के बारे में बात करेंगे जिससे इस वर्ल्ड कप में भारतिय फैंस को काफी उमीदे रहेंगे"
},
{
"timestamp": [
5.42,
13.46
],
"text": " इस खिलाड़ी के बारे में मैं आपको बस इतना बता दूं कि ये वो खिलाड़ी हैं जिसने इंटरनेशनल क्रिकेट में अपनी एंट्री का एलान औस्ट्रेलिया के गेंज जोरदार प्रदर्शन से किया"
},
{
"timestamp": [
13.62,
19.52
],
"text": " आज से वर्ल्ड कप के दारान हर रोज हम आपको मिलते रहेंगे EAM Cricket World Cup 2007 अपडेट में"
}
]
},
"resultTranslation": {
"text": " Now we will talk about a player who will be a big fan of Indian fans in this World Cup. I will tell you that he is a player who has made his entry in international cricket with a strong Australian against. From today during the World Cup, we will meet you every day in EAM Cricket World Cup 2007 Update.",
"chunks": [
{
"timestamp": [
0.0,
5.6
],
"text": " Now we will talk about a player who will be a big fan of Indian fans in this World Cup."
},
{
"timestamp": [
5.6,
13.6
],
"text": " I will tell you that he is a player who has made his entry in international cricket with a strong Australian against."
},
{
"timestamp": [
13.6,
19.64
],
"text": " From today during the World Cup, we will meet you every day in EAM Cricket World Cup 2007 Update."
}
]
},
"resultClassifier": [
{
"label": "POSITIVE",
"score": 0.999645471572876
}
]
}
19 changes: 18 additions & 1 deletion documentation/GETTING_STARTED.MD
View file
Original file line number Diff line number Diff line change
Expand Up @@ -4,21 +4,38 @@ These instructions will get you a copy of the project up and running on your loc

~~~shell
python --version
python3 --version
~~~

~~~shell
pip --version
~~~

~~~shell
python -c "import torch; print(torch.__version__)"
pip3 show torch
~~~

~~~shell
pip3 list | grep tensorflow
pip list | grep tensorflow
~~~

~~~shell
pip list
~~~

---

~~~shell
python -m venv .env

source .env/bin/activate
~~~
~~~

pip install transformers
pip install torch
pip install datasets
pip install accelerate
pip install librosa
sudo apt install ffmpeg
18 changes: 9 additions & 9 deletions documentation/TECHNOLOGY_STACK.MD
View file
Original file line number Diff line number Diff line change
Expand Up @@ -2,13 +2,13 @@

### Overview

|Technology |Description |
|--------------------------|--------------------|
|PyTorch/Tensorflow | |
|Python | |
|Transformers | |
|openai/whisper-large-v3 | |



|Technology | Description |
|--------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
|[PyTorch](https://pytorch.org/) | A machine learning framework based on the Torch library, used for applications such as computer vision and natural language processing, originally developed by Meta AI and now part of the Linux Foundation umbrella. |
|[Python](https://www.python.org/) | A high-level, general-purpose programming language |
|[Transformers](https://huggingface.co/docs/transformers/en/index) | It provides APIs and tools to easily download and train state-of-the-art pretrained models |
|[openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) | A pre-trained model for automatic speech recognition (ASR) and speech translation.a pre-trained model for automatic speech recognition (ASR) and speech translation. |
|[Accelerate](https://github.com/huggingface/accelerate) | A library that enables the same PyTorch code to be run across any distributed configuration |
|[librosa](https://librosa.org/doc/latest/index.html) | A python package for music and audio analysis. |
|[FFmpeg](https://ffmpeg.org/) | A suite of libraries and programs for handling video, audio, and other multimedia files and streams |

0 comments on commit a63d25a

Please sign in to comment.