change interface; todo: update changelog

Author: eff-kay

examples/pipeline.py

@@ -0,0 +1,36 @@
+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.ASSEMBLY
+service_provider_summarize = TextSummarization.ServiceProvider.OPENAI
+service_provider_text_to_speech = TextToSpeech.ServiceProvider.AWS
+
+# 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
+)
+
+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.pyexamples/speech_to_text.py renamed to 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.py examples/summarize_async.pyexamples/summarize.py renamed to examples/summarize_async.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/summarize_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}")

examples/text_to_speech.py examples/text_to_speech_async copy.pyexamples/text_to_speech.py renamed to examples/text_to_speech_async copy.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 +37,30 @@ 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)
+
+        import pdb
+        pdb.set_trace()
+        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
+    

examples/text_to_speech_sync.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/speech_to_text.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