change interface; todo: update changelog

CHANGELOG.md

@@ -2,7 +2,12 @@
 All notable changes to `slashml-python-client` aka `slashml` will be documented in this file.
 This project adheres to [Semantic Versioning](https://semver.org/).
 
-## 0.0.4 - 2023-05-08
+## 0.0.6 - 2023-05-11
+
+### Changed
+- Combined submit_job and status check into a single function call.
+
+## 0.0.5 - 2023-05-08
 
 ### Added
 - SpeechToText service

README.md

@@ -16,92 +16,56 @@ pip install slashml
 
 ```python
 from slashml import TextToSpeech
-import time
 
-# Initialize model
-model = TextToSpeech()
-
-service_provider = TextToSpeech.ServiceProvider.GOOGLE
-
-# Submit speechification request
-job = model.speechify(text="To be or not to be, that is the question!", service_provider=service_provider)
-
-assert job.status != "ERROR", f"{job}"
-print(f"Job ID: {job.id}")
+# Replace `API_KEY` with your SlasML API token. This example still runs without
+# the API token but usage will be limited
+model = TextToSpeech(api_key=None)
 
-# check job status
-response = model.status(job.id, service_provider=service_provider)
+input_text = "to be or not to be, that is the question!"
 
-while response.status == "IN_PROGRESS":
-    time.sleep(5)
-    response = model.status(job.id, service_provider=service_provider)
-    print(f"Response = {response}. Retrying in 5 seconds")
+# Submit request
+job = model.execute(text=input_text, service_provider=TextToSpeech.ServiceProvider.AWS)
 
-print(response)
+print (f"\n\n\n You can access the audio file here: {job.audio_url}")
 ```
 
 #### Transcribe an audio file
 <!-- write a code snippet in the minimum number of lines  -->
 
 ```python
 from slashml import SpeechToText
-import time
-
-# Initialize model
-model = SpeechToText()
-
-service_provider = SpeechToText.ServiceProvider.WHISPER
-# Submit transcription request
-job = model.transcribe(upload_url="https://slashml.s3.ca-central-1.amazonaws.com/c7d38026-3ab4-4a04-ad9e-b6679ab79a87", service_provider=service_provider)
 
-assert job.status != "ERROR", f"{job}"
-print(f"Job ID: {job.id}")
+# Replace `API_KEY` with your SlasML API token. This example still runs without
+# the API token but usage will be limited
+model = SpeechToText(api_key=None)
 
-# check job status
-response = model.status(job.id, service_provider=service_provider)
+response = model.execute(
+    upload_url='https://slashml.s3.ca-central-1.amazonaws.com/695c711f-9f5d-4ff1-ae4f-4439842eef5f', 
+    service_provider=SpeechToText.ServiceProvider.WHISPER
+)
 
-while response.status == "IN_PROGRESS":
-    time.sleep(5)
-    response = model.status(job.id, service_provider=service_provider)
-    print(f"Response = {response}. Retrying in 5 seconds")
+print(f"\n\n\n\nTranscription = {response.transcription_data.transcription}")
 
-print(response)
 ```
 
 #### Summarize a text input
 <!-- write a code snippet in the minimum number of lines  -->
 
 ```python
 from slashml import TextSummarization
-import time
 
-# Initialize model
-model = TextSummarization()
-
-service_provider = TextSummarization.ServiceProvider.OPENAI
-
-# Submit summariztion request
-job = model.summarize(text="There are of course kinds of thinking that can be done without writing. If you don't need to go too deeply into a problem, you can solve it without writing. If you're thinking about how two pieces of machinery should fit together, writing about it probably won't help much. And when a problem can be described formally, you can sometimes solve it in your head. But if you need to solve a complicated, ill-defined problem, it will almost always help to write about it. Which in turn means that someone who's not good at writing will almost always be at a disadvantage in solving such problems.", service_provider=service_provider)
+model = TextSummarization(api_key=None)
 
-assert job.status != "ERROR", f"{job}"
-print(f"Job ID: {job.id}")
+input_text = """A good writer doesn't just think, and then write down what he thought, as a sort of transcript. A good writer will almost always discover new things in the process of writing. And there is, as far as I know, no substitute for this kind of discovery. Talking about your ideas with other people is a good way to develop them. But even after doing this, you'll find you still discover new things when you sit down to write. There is a kind of thinking that can only be done by writing."""
 
-# check job status
-response = model.status(job.id, service_provider=service_provider)
+response = model.execute(text=input_text, service_provider=TextSummarization.ServiceProvider.OPENAI)
 
-while response.status == "IN_PROGRESS":
-    time.sleep(5)
-    response = model.status(job.id, service_provider=service_provider)
-    print(f"Response = {response}. Retrying in 5 seconds")
+print(f"Summary = {response.summarization_data}")
 
-print(response)
 ```
 
 
-
-
 ### View the list of service providers available
-
 ```python
 from slashml import TextToSpeech
 

examples/pipeline_podcast_summary.py

@@ -0,0 +1,37 @@
+from slashml import SpeechToText, TextSummarization, TextToSpeech
+
+# Replace `API_KEY` with your SlasML API token. This example still runs without
+# the API token but usage will be limited
+API_KEY = "a7011983a0f3d64ee113317b1e36f8e5bf56c14a"
+
+service_provider_speech_to_text = SpeechToText.ServiceProvider.WHISPER
+service_provider_summarize = TextSummarization.ServiceProvider.OPENAI
+service_provider_text_to_speech = TextToSpeech.ServiceProvider.GOOGLE
+
+# 10 minute audio file already uploaded
+uploaded_url = (
+    "https://slashml.s3.ca-central-1.amazonaws.com/fda70f6a-6057-4541-adf1-2cf4f4182929"
+)
+
+transcribe = SpeechToText(api_key=API_KEY)
+summarize = TextSummarization(api_key=API_KEY)
+speechify = TextToSpeech(api_key=API_KEY)
+
+
+response = transcribe.execute(
+    upload_url=uploaded_url, service_provider=service_provider_speech_to_text
+)
+
+print('starting pipeline, the first response might take 10 secs')
+transcribed_text = response.transcription_data.transcription
+print (f"Transcribed Text = {transcribed_text}")
+
+response_summarize = summarize.execute(transcribed_text, service_provider_summarize)
+
+summary = response_summarize.summarization_data
+
+print (f"Summarized Text = {summary}")
+
+response = speechify.execute(summary, service_provider_text_to_speech)
+
+print (f"\n\n\n You can access the audio file here: {response.audio_url}")

examples/speech_to_text.py → examples/speech_to_text_async.py

@@ -1,6 +1,7 @@
 from slashml import SpeechToText
 import time
 
+# update the examples using execute from the SpeechToText class
 
 def speech_to_text(audio_filepath, service_provider, api_key):
     # Initialize model

examples/speech_to_text_sync.py

@@ -0,0 +1,24 @@
+from slashml import SpeechToText
+
+# Replace `API_KEY` with your SlasML API token. This example still runs without
+# the API token but usage will be limited
+API_KEY = None
+service_provider = SpeechToText.ServiceProvider.AWS
+audio_filepath = "test.mp3"
+
+# Find all the service providers that we support by running the choices() method
+print(f"Available providers: {SpeechToText.ServiceProvider.choices()}")
+print(f"Selected provider: {service_provider}")
+
+
+model = SpeechToText(api_key=API_KEY)
+
+# Upload audio
+uploaded_file = model.upload_audio(audio_filepath)
+print(f"file uploaded: {uploaded_file}")
+
+response = model.execute(
+    upload_url=uploaded_file["upload_url"], service_provider=service_provider
+)
+
+print(f"\n\n\n\nTranscription = {response.transcription_data.transcription}")

examples/summarize_sync.py

@@ -0,0 +1,19 @@
+from slashml import TextSummarization
+
+
+# Replace `API_KEY` with your SlasML API token. This example still runs without
+# the API token but usage will be limited
+API_KEY = "a7011983a0f3d64ee113317b1e36f8e5bf56c14a"
+service_provider = TextSummarization.ServiceProvider.OPENAI
+
+input_text = """A good writer doesn't just think, and then write down what he thought, as a sort of transcript. A good writer will almost always discover new things in the process of writing. And there is, as far as I know, no substitute for this kind of discovery. Talking about your ideas with other people is a good way to develop them. But even after doing this, you'll find you still discover new things when you sit down to write. There is a kind of thinking that can only be done by writing."""
+
+model = TextSummarization(api_key=API_KEY)
+
+# Find all the service providers that we support by running the choices() method
+print(f"Available providers: {TextSummarization.ServiceProvider.choices()}")
+print(f"Selected provider: {service_provider}")
+
+response = model.execute(text=input_text, service_provider=service_provider)
+
+print(f"Summary = {response.summarization_data}")

examples/text_to_speech_sync.py

@@ -0,0 +1,18 @@
+from slashml import TextToSpeech
+
+# Replace `API_KEY` with your SlasML API token. This example still runs without
+# the API token but usage will be limited
+API_KEY = None
+service_provider = TextToSpeech.ServiceProvider.AWS
+input_text = "To be or not to be, that is the question!"
+
+# Find all the service providers that we support by running the choices() method
+print(f"Available providers: {TextToSpeech.ServiceProvider.choices()}")
+print(f"Selected provider: {service_provider}")
+
+model = TextToSpeech(api_key=API_KEY)
+
+# Submit request
+job = model.execute(text=input_text, service_provider=service_provider)
+
+print (f"\n\n\n You can access the audio file here: {job.audio_url}")

setup.py

@@ -14,7 +14,7 @@ def read_req_file(req_type):
 
 setup(
     name="slashml",
-    version="0.0.5",
+    version="0.0.6",
     url="https://slashml.com/",
     author="eff-kay",
     author_email="faiizan14@gmail.com",

slashml/speech_to_text.py

@@ -1,4 +1,5 @@
 import requests
+import time
 from enum import Enum
 from .utils import generateURL, baseUrl, generateHeaders, formatResponse, getTaskStatus
 
@@ -25,7 +26,7 @@ def upload_audio(self, file_location: str):
         response = requests.post(url, headers=self._headers, files=files)
         return formatResponse(response)
 
-    def transcribe(self, upload_url: str, service_provider: ServiceProvider):
+    def submit_job(self, upload_url: str, service_provider: ServiceProvider):
         url = generateURL(self._base_url, "jobs")
         payload = {
             "uploaded_audio_url": upload_url,
@@ -36,3 +37,28 @@ def transcribe(self, upload_url: str, service_provider: ServiceProvider):
 
     def status(self, job_id: str, service_provider: ServiceProvider):
         return getTaskStatus(self._base_url, self._headers, job_id, service_provider)
+
+    def execute(self, upload_url: str, service_provider: ServiceProvider):
+        url = generateURL(self._base_url, "jobs")
+
+        payload = {
+            "uploaded_audio_url": upload_url,
+            "service_provider": service_provider.value,
+        }
+
+        response = requests.post(url, headers=self._headers, data=payload)
+        job = formatResponse(response)
+
+        assert job.status != "ERROR", f"{job}"
+        print(f"Got Job ID: {job.id}")
+
+        # check job status
+        response = getTaskStatus(self._base_url, self._headers, job.id, service_provider)
+
+        while response.status == "IN_PROGRESS":
+            time.sleep(5)
+            response = getTaskStatus(self._base_url, self._headers, job.id, service_provider)
+            print(f"Response = {response}. Retrying in 5 seconds")
+
+        return response
+

slashml/text_summarization.py

@@ -1,4 +1,5 @@
 import requests
+import time
 from enum import Enum
 from .utils import generateURL, baseUrl, generateHeaders, formatResponse, getTaskStatus
 
@@ -17,11 +18,31 @@ def choices(cls):
     def __init__(self, api_key: str = None):
         self._headers = generateHeaders(api_key)
 
-    def summarize(self, text: str, service_provider: ServiceProvider):
+    def submit_job(self, text: str, service_provider: ServiceProvider):
         url = generateURL(self._base_url, "jobs")
         payload = {"text": [text], "service_provider": service_provider.value}
         response = requests.post(url, headers=self._headers, data=payload)
         return formatResponse(response)
 
     def status(self, job_id: str, service_provider: ServiceProvider):
         return getTaskStatus(self._base_url, self._headers, job_id, service_provider)
+
+    def execute(self, text: str, service_provider: ServiceProvider):
+        url = generateURL(self._base_url, "jobs")
+        payload = {"text": [text], "service_provider": service_provider.value}
+        response = requests.post(url, headers=self._headers, data=payload)
+        job = formatResponse(response)
+
+        assert job.status != "ERROR", f"{job}"
+        print(f"Got Job ID: {job.id}")
+
+        # check job status
+        response = getTaskStatus(self._base_url, self._headers, job.id, service_provider)
+
+        while response.status == "IN_PROGRESS":
+            time.sleep(5)
+            response = getTaskStatus(self._base_url, self._headers, job.id, service_provider)
+            print(f"Response = {response}. Retrying in 5 seconds")
+
+        return response
+

slashml/text_to_speech.py

@@ -1,4 +1,6 @@
 import requests
+import time
+
 from enum import Enum
 from .utils import generateURL, baseUrl, generateHeaders, formatResponse, getTaskStatus
 
@@ -18,11 +20,30 @@ def choices(cls):
     def __init__(self, api_key: str = None):
         self._headers = generateHeaders(api_key)
 
-    def speechify(self, text: str, service_provider: ServiceProvider):
+    def submit_job(self, text: str, service_provider: ServiceProvider):
         url = generateURL(self._base_url, "jobs")
         payload = {"text": text, "service_provider": service_provider.value}
         response = requests.post(url, headers=self._headers, data=payload)
         return formatResponse(response)
 
     def status(self, job_id: str, service_provider: ServiceProvider):
         return getTaskStatus(self._base_url, self._headers, job_id, service_provider)
+
+    def execute(self, text: str, service_provider: ServiceProvider):
+        url = generateURL(self._base_url, "jobs")
+        payload = {"text": text, "service_provider": service_provider.value}
+        response = requests.post(url, headers=self._headers, data=payload)
+        job = formatResponse(response)
+
+        assert job.status != "ERROR", f"{job}"
+        print(f"Got Job ID: {job.id}")
+
+        # check job status
+        response = getTaskStatus(self._base_url, self._headers, job.id, service_provider)
+
+        while response.status == "IN_PROGRESS":
+            time.sleep(5)
+            response = getTaskStatus(self._base_url, self._headers, job.id, service_provider)
+            print(f"Response = {response}. Retrying in 5 seconds")
+
+        return response