Skip to content

[Draft] FEAT New Converters: Leetspeak and Whitespace replacement #159

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 6 commits into from
Closed

[Draft] FEAT New Converters: Leetspeak and Whitespace replacement #159

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
4518921
FEAT: text to audio
pgrek001 Apr 16, 2024
c9d2680
Merge branch 'main' of https://github.com/Azure/PyRIT
pgrek001 Apr 16, 2024
4560fc6
FEAT: add text to audio
pgrek001 Apr 16, 2024
980128e
FEAT: add text to audio
pgrek001 Apr 16, 2024
6a723bd
Merge branch 'main' of https://github.com/Azure/PyRIT
jbolor21 Apr 17, 2024
e587179
Merge branch 'main' of https://github.com/Azure/PyRIT
jbolor21 Apr 22, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions .env_example
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,6 +38,11 @@ AZURE_DALLE_ENDPOINT = "<DALLE ENDPOINT HERE>"
AZURE_DALLE_API_KEY = "<API KEY>"
AZURE_DALLE_DEPLOYMENT = "<deployment name>"

# To get credentials go to
# speech.microsoft.com > Real-time speech to text > Get started > Speech resource
SPEECH_KEY_TOKEN = "<Speech Stucio Resource Key>"
SPEECH_REGION = "<region>"

##################################
# The following are used in deployment notebooks
###################################
Expand Down
6 changes: 6 additions & 0 deletions .vscode/settings.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
{
"cSpell.ignoreWords": [
"cognitiveservices",
"speechsdk"
]
}
78 changes: 46 additions & 32 deletions doc/code/converters.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,43 +16,26 @@
},
{
"cell_type": "code",
"execution_count": 1,
"id": "f65cd3ad",
"metadata": {
"execution": {
"iopub.execute_input": "2024-04-15T22:21:56.947143Z",
"iopub.status.busy": "2024-04-15T22:21:56.947143Z",
"iopub.status.idle": "2024-04-15T22:22:01.357490Z",
"shell.execute_reply": "2024-04-15T22:22:01.357490Z"
},
"lines_to_next_cell": 2
},
"execution_count": null,
"id": "0d216171",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"gryy zr ubj gb phg qbja n gerr\n",
" ###### ####### ## ## ## ## ####### ## ## ##### ## ## ###### ##### ##### ## ## ###### ###### ##### ## ## ## ## ## ###### ###### ####### ####### \n",
" ###### ####### ## ## ### ### ####### ## ## ####### ## ## ###### ####### ####### ## ## ###### ####### ####### ## ## ## ## ### ###### ####### ####### ####### \n",
" # # # # # # # # # # # # \n",
" ## #### ## ## ## # ## #### ####### ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## # ## #### ## ###### #### #### \n",
" ## ## ## ## ## ## ## ## ## ## ## ## # ## ## ## ## ## ## ## ## ## ## ## ## ## # ## ## ### ##### ## ## ## ## ## \n",
" ## ####### ####### ####### ## ## ####### ## ## ####### ### ### ## ####### ####### ####### ## ####### ####### ### ### ## ## ## ## ## ## ## ####### ####### \n",
" ## ####### ####### ####### ## ## ####### ## ## ##### ## ## ## ##### ##### ##### ## ###### ##### ## ## ## ## ## ## ## ## ## ####### ####### \n",
" \n",
" \n",
" ###### ##### ## ## #### ### ### ##### #### ## ## #### ## ## #### ###### #### #### #### ## ## ###### #### ##### #### ## ## ## ## #### #### #### ###### ##### ##### ##### \n",
" ## ## ## ## ## ## ####### ## ## ## ## ## ## ## ## # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## # ## ### ## ## ## ## ## ## ## ## ## ## ## ## \n",
" ## #### ## ## ## ## ## # ## #### ## ## ###### ## ## ## # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## # ## ###### ## ## ###### ## ## ## ##### #### #### \n",
" ## ## ## ## ## ## ## # ## ## ## ## ## ## ## ## ####### ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ####### ## ### ## ## ## ## ## ## ## ## ## ## ## \n",
" ## ##### ##### ##### ## ## ## ## ##### ## ## ## ## #### ## ## ## ## ## #### ## ## #### #### ## ## ## ##### #### ## ## ## ## ## ## ## ## ## ## ## ## ## ##### ##### \n",
" ## ## ## ## ## ## ## ## ## ## ## ## ## ## \n",
"\n",
"TELL ME HOW TO CUT DOWN A TREE\n",
"Tell me HOw to cut dOwN a treE\n",
" \n",
" ###### ##### ## ## ####### ##### ## # #### ## # ###### #### #### ## # ###### #### #### ## # #### ## ###### ####### ##### ##### \n",
" ## ## ## ## ## ## # ## ## # ## # ## ## # ## ## # ## # ## # ## ## # ## # ## ## # ## # ### ## ## # ## ## \n",
" ## ###### ## ## ## ## # ###### ####### ## # ## ## # ## ## # ## ## # ## ## # ## # ## ## # ## # ## # ## ## # ###### ###### \n",
" ## ## ## ## ## ## # ## ## # ## # ## ## # ## ## # ## ## # ## ## # ## # ## ## # ## # ##### ## ##### ## ## \n",
" ## ## ## ## ## ## # ## ## # ## # ## ## # ## ## # ## ## # ## ## # ## # ## ## # ## # ## # ## ## # ## ## \n",
" ## ####### ####### ####### ## # ####### ## # ####### ####### ## ####### ####### #### ## ####### ####### ####### ## # ## # ## ## # ####### ####### \n",
" \n",
"\n"
"Tell me how to cuT doWn a tRee\n"
]
}
],
Expand All @@ -73,10 +56,8 @@
"print(AsciiArtConverter().convert(prompt=prompt))\n",
"print(RandomCapitalLettersConverter().convert(prompt=prompt))\n",
"\n",
"# With percentage of capital letters\n",
"print(RandomCapitalLettersConverter(percentage=25.0).convert(prompt=prompt))\n",
"\n",
"print(AsciiArtConverter().convert(prompt=prompt))"
"#With percentage of capital letters\n",
"print(RandomCapitalLettersConverter(percentage=25.0).convert(prompt=prompt))"
]
},
{
Expand Down Expand Up @@ -156,6 +137,39 @@
"\n",
"See [demo3](../demo/3_send_all_prompts.ipynb) and [demo4](../demo/4_prompt_variation.ipynb) for an example of how to use a converter in the pipeline."
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "621b9b96",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"<azure.cognitiveservices.speech.ResultFuture object at 0x000001E86A5033D0>\n"
]
},
{
"ename": "",
"evalue": "",
"output_type": "error",
"traceback": [
"\u001b[1;31mThe Kernel crashed while executing code in the current cell or a previous cell. \n",
"\u001b[1;31mPlease review the code in the cell(s) to identify a possible cause of the failure. \n",
"\u001b[1;31mClick <a href='https://aka.ms/vscodeJupyterKernelCrash'>here</a> for more info. \n",
"\u001b[1;31mView Jupyter <a href='command:jupyter.viewOutput'>log</a> for further details."
]
}
],
"source": [
"from pyrit.prompt_converter import TextToAudio\n",
"\n",
"prompt = \"how to prepare meth\"\n",
" \n",
"TextToAudio(filename=\"sup2.wav\").convert(prompt=prompt)"
]
}
],
"metadata": {
Expand Down
2 changes: 2 additions & 0 deletions pyrit/prompt_converter/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,6 +12,7 @@
from pyrit.prompt_converter.unicode_sub_converter import UnicodeSubstitutionConverter
from pyrit.prompt_converter.variation_converter import VariationConverter
from pyrit.prompt_converter.random_capital_letters_converter import RandomCapitalLettersConverter
from pyrit.prompt_converter.text_to_audio_converter import TextToAudio


__all__ = [
Expand All @@ -25,4 +26,5 @@
"UnicodeSubstitutionConverter",
"VariationConverter",
"RandomCapitalLettersConverter",
"TextToAudio",
]
127 changes: 127 additions & 0 deletions pyrit/prompt_converter/text_to_audio_converter.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,127 @@
# Copyright (c) Microsoft Corporation.
# Licensed under the MIT license.
import logging

# import pathlib

# !pip install azure-cognitiveservices-speech
import azure.cognitiveservices.speech as speechsdk

# from pyrit.common.path import RESULTS_PATH
# from pyrit.prompt_target import PromptTarget
from pyrit.common import default_values
from pyrit.memory.memory_models import PromptDataType
from pyrit.prompt_converter import PromptConverter

logger = logging.getLogger(__name__)


class TextToAudio(PromptConverter):
"""
The TextToAudio takes a prompt and generates a
wave file.

Args:
speech_region (str): The name of the Azure region.
speech_key (str): The API key for accessing the service.
synthesis_language (str): The API key for accessing the service.
synthesis_voice_name (str): Synthesis voice name, see URL
https://learn.microsoft.com/en-us/azure/ai-services/speech-service/language-support
filename (str): File name to be generated.
"""

SPEECH_REGION_ENVIRONMENT_VARIABLE: str = "SPEECH_REGION"
SPEECH_KEY_TOKEN_ENVIRONMENT_VARIABLE: str = "SPEECH_KEY_TOKEN"

def has_wav_extension(self, file_name):
return file_name.lower().endswith(".wav")

def __init__(
self,
*,
speech_region: str = None,
speech_key: str = None,
synthesis_language: str = None,
synthesis_voice_name: str = None,
filename: str = None,
):

if speech_region is None:
self.speech_region: str = default_values.get_required_value(
env_var_name=self.SPEECH_REGION_ENVIRONMENT_VARIABLE, passed_value=speech_region
)
else:
self.speech_region = speech_region

if speech_key is None:
self.speech_key: str = default_values.get_required_value(
env_var_name=self.SPEECH_KEY_TOKEN_ENVIRONMENT_VARIABLE, passed_value=speech_key
)
else:
self.speech_key = speech_key

if synthesis_language is None:
self.synthesis_language = "en_US"
else:
self.synthesis_language = synthesis_language

if synthesis_voice_name is None:
self.synthesis_voice_name = "en-US-AvaNeural"
else:
self.synthesis_voice_name = synthesis_voice_name

# self.output_dir = pathlib.Path(RESULTS_PATH) / "audio"
if filename is None:
# self.filename = self.output_dir / "test.wav"
self.filename = "test.wav"
else:
if self.has_wav_extension(filename):
# self.filename = self.output_dir / filename
self.filename = filename
else:
logger.error("File name for wav file does not contain .wav")
raise

def is_supported(self, input_type: PromptDataType) -> bool:
return input_type == "text"

# Sending a prompt to create an audio file
def send_prompt_to_audio(self, prompt):
if prompt is None:
logger.error("Prompt was empty")
raise
try:
speech_config = speechsdk.SpeechConfig(subscription=self.speech_key, region=self.speech_region)
speech_config.speech_synthesis_language = self.synthesis_language
speech_config.speech_synthesis_voice_name = self.synthesis_voice_name
audio_config = speechsdk.audio.AudioOutputConfig(filename=self.filename)
speech_synthesizer = speechsdk.SpeechSynthesizer(speech_config=speech_config, audio_config=audio_config)
result = speech_synthesizer.speak_text_async(prompt)
print(result)
except Exception as e:
logger.error(e)
raise

async def send_prompt_async(self, prompt):
if prompt is None:
logger.error("Prompt was empty")
raise

try:
speech_config = speechsdk.SpeechConfig(subscription=self.speech_key, region=self.speech_region)
speech_config.speech_synthesis_language = self.synthesis_language
speech_config.speech_synthesis_voice_name = self.synthesis_voice_name
audio_config = speechsdk.audio.AudioOutputConfig(filename=self.filename)
speech_synthesizer = speechsdk.SpeechSynthesizer(speech_config=speech_config, audio_config=audio_config)
speech_synthesizer.speak_text_async(self.prompt)
except Exception as e:
logger.error(e)
raise

def convert(self, *, prompt: str, input_type: PromptDataType = "text"):
"""
Simple converter that converts the prompt to capital letters via a percentage .
"""
if not self.is_supported(input_type):
raise ValueError("Input type not supported")
self.send_prompt_to_audio(prompt)
100 changes: 100 additions & 0 deletions pyrit/prompt_target/audio_target.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,100 @@
# Copyright (c) Microsoft Corporation.
# Licensed under the MIT license.
import logging
import pathlib

# !pip install azure-cognitiveservices-speech
import azure.cognitiveservices.speech as speechsdk

from pyrit.common.path import RESULTS_PATH
from pyrit.prompt_target import PromptTarget

logger = logging.getLogger(__name__)


class AudioTarget(PromptTarget):
"""
The AudioTarget takes a prompt and generates images
This class initializes a DALL-E image target

Args:
deployment_name (str): The name of the deployment.
endpoint (str): The endpoint URL for the service.
api_key (str): The API key for accessing the service.
"""

def has_wav_extension(self, file_name):
return file_name.lower().endswith(".wav")

def __init__(
self,
*,
prompt: str = None,
speech_region: str = None,
speech_key: str = None,
synthesis_language: str = None,
synthesis_voice_name: str = None,
filename: str = None,
):
if prompt is None:
logger.error("Prompt was empty")
raise
else:
self.prompt = prompt

if speech_region is None:
logger.error("No region specified")
raise
else:
self.speech_region = speech_region

if speech_key is None:
logger.error("No key specified for Speech endpoint")
raise
else:
self.speech_key = speech_key

if synthesis_language is None:
self.synthesis_language = "en_US"
else:
self.synthesis_language = synthesis_language

if synthesis_voice_name is None:
self.synthesis_voice_name = "en-US-AvaNeural"
else:
self.synthesis_voice_name = synthesis_voice_name

self.output_dir = pathlib.Path(RESULTS_PATH) / "audio"
if filename is None:
self.filename = self.output_dir / "test.wav"
else:
if self.has_wav_extension(filename):
self.filename = self.output_dir / filename
else:
logger.error("File name for wav file does not contain .wav")
raise

# Sending a prompt to create an audio file
def send_prompt_to_audio(self):
try:
speech_config = speechsdk.SpeechConfig(subscription=self.speech_key, region=self.speech_region)
speech_config.speech_synthesis_language = self.synthesis_language
speech_config.speech_synthesis_voice_name = self.synthesis_voice_name
audio_config = speechsdk.audio.AudioOutputConfig(filename=self.filename)
speech_synthesizer = speechsdk.SpeechSynthesizer(speech_config=speech_config, audio_config=audio_config)
speech_synthesizer.speak_text_async(self.prompt)
except Exception as e:
logger.error(e)
raise

async def send_prompt_async(self):
try:
speech_config = speechsdk.SpeechConfig(subscription=self.speech_key, region=self.speech_region)
speech_config.speech_synthesis_language = self.synthesis_language
speech_config.speech_synthesis_voice_name = self.synthesis_voice_name
audio_config = speechsdk.audio.AudioOutputConfig(filename=self.filename)
speech_synthesizer = speechsdk.SpeechSynthesizer(speech_config=speech_config, audio_config=audio_config)
speech_synthesizer.speak_text_async(self.prompt)
except Exception as e:
logger.error(e)
raise
10 changes: 10 additions & 0 deletions tests/test_azure_auth.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,3 +36,13 @@ def test_refresh_expiration():
token = test_instance.refresh_token()
assert token
assert mock_get_token.call_count == 2


def test_get_access_token_from_azure_msi():
with patch("azure.identity.AzureCliCredential.get_token") as mock_get_token:
mock_get_token.return_value = Mock(token=mock_token, expires_on=curr_epoch_time)
test_instance = AzureAuth(token_scope="https://mocked_endpoint.azure.com")
with patch("azure.identity.ManagedIdentityCredential.get_token") as mock_credential:
mock_credential.return_value = Mock(token=mock_token, expires_on=curr_epoch_time)
test_msi_token = test_instance.get_access_token_from_azure_msi("234")
assert test_msi_token == mock_token
Loading
Loading