updated sdk with service_provider

Author: JJneid

.DS_Store

@@ -1,3 +1,56 @@
-# slashml-python-client
+# SLASHML Python client
 
-to be updated to guide users
+This is a Python client for SLASHML. It lets you run transcription jobs from your Python code or Jupyter notebook.
+
+Grab your token from [https://www.slashml.com/dashboard/](>settings> new api key) and authenticate by setting it as an environment variable (or when you initialize the service, see examples):
+```
+export SLASHML_API_KEY=[token]
+```
+
+-- update from this part, include examples, sign up, token, service providers, type of servies, benchmarking, link to pricing, Tutorial examples/examples
+
+
+SDK for SlashML documentation:
+- methods: upload_audio, transcribe, status 
+
+Steps to Integrate
+1 - (Optional) Upload files where the data points to your audio file
+```
+# call the class
+speect_to_text = speechtotext.SpeechToText()
+file_location="path/to/your/file.mp3"
+# when
+API_KEY="SLASH_ML_API_KEY"
+model_choice="assembly"
+result_upload = speect_to_text.upload_audio(file_location,API_KEY, model_choice)
+print(result_upload)
+```
+Save the upload_url. You can use this url link in the rest of the calls.
+
+
+2- Submit your audio file for transcription
+```
+upload_url=upload_url # you can skip step 1 and just input the accessible link of your # file)
+
+result_transcribe = speect_to_text.transcribe(upload_url,API_KEY, model_choice)
+
+print(result_transcribe)
+```
+Save the id in the response object.
+
+
+3 - Check the status and get the text result of the transcription
+```
+job_id= id
+result_status = speect_to_text.status(job_id,API_KEY, model_choice=model_choice)
+
+### get the full details of the result
+print(result_status)
+### get the text reulst only
+print(json.loads(result)["text"])
+```
+
+
+et voilà, Next steps:
+- pip install slashml
+- add SLASH_API_KEY to sys path

README.md

@@ -2,9 +2,10 @@
 import requests
 import os
 from pathlib import Path
+import json
 
 class SpeechToText:
-    # TODO: make sure this is dynamic
+
     SLASHML_BASE_URL = 'https://api.slashml.com/speech-to-text/v1'
     SLASHML_UPLOAD_URL = SLASHML_BASE_URL+'/upload/'
     SLASHML_TRANSCRIPT_URL = SLASHML_BASE_URL+'/transcribe/'
@@ -13,46 +14,42 @@ class SpeechToText:
 
     HEADERS:dict = {}
     ## add the api key to sys path envs 
-    def __init__(self) -> None:
-        self.HEADERS = {'authorization': os.environ.get('SLASHML_API_KEY')}
+    def __init__(self,SLASHML_API_KEY) -> None:
+        self.HEADERS = {'authorization': os.environ.get('SLASHML_API_KEY')},
+        self.API_KEY=SLASHML_API_KEY
 
-    def upload_audio(self, file_location:str, API_KEY: str, model_choice: str ):
-        # here we can also add the service? assemblyai, aws, gcp?
+    def upload_audio(self, file_location:str):
 
-        token="Token {0}".format(API_KEY)
+        token="Token {0}".format(self.API_KEY)
         headers = {'authorization': token}
-        payload = { 'model':model_choice }
+
         files=[
         ('audio',('test_audio.mp3',open(file_location,'rb'),'audio/mpeg'))
         ]
-        #import pdb
-        #pdb.set_trace()
         response = requests.post(self.SLASHML_UPLOAD_URL,
-                                headers=headers,
-                                data=payload,files=files)
-        return response.text
+                                headers=headers,files=files)
+        
+        return json.loads(response.text)["upload_url"]
+
+    def transcribe(self,upload_url:str,  service_provider: str ):
 
-    def transcribe(self,upload_url:str, API_KEY: str, model_choice: str ):
-        # here we can add more model params
         transcript_request = {'audio_url': upload_url}
-        token="Token {0}".format(API_KEY)
+        token="Token {0}".format(self.API_KEY)
         headers = {'authorization': token}
         payload = {
         "uploaded_audio_url": upload_url,
-        "model": model_choice
+        "service_provider": service_provider
         }
-
         response = requests.post(self.SLASHML_TRANSCRIPT_URL, headers=headers, data=payload)
-        return response.text
+        return json.loads(response.text)["id"]
 
-    def status(self,job_id:str, API_KEY: str, model_choice: str ):
-        token="Token {0}".format(API_KEY)
+    def status(self,job_id:str, service_provider: str ):
+        token="Token {0}".format(self.API_KEY)
         headers = {'authorization': token}
         payload = {
-        "model": model_choice
+        "service_provider": service_provider
         }
-    #import pdb
-       # pdb.set_trace()        
+       
         response = requests.get(self.SLASHML_TRANSCRIPT_STATUS_URL(job_id) , headers=headers, data=payload)
-        return response.text
+        return json.loads(response.text)
 

SDK_v1/.DS_Store

@@ -0,0 +1,26 @@
+import speechtotext
+
+#initialize SLASHML
+API_KEY="1bd15e1c161ff6d4db2ea1d661d7468b4fa61ca9"
+speect_to_text = speechtotext.SpeechToText(API_KEY)
+# optional local file to upload, if not an accessible url
+file_location="/Users/JJneid/Desktop/SlashMl/s2t_experiments/api_tests/test_audio_1.mp3" 
+
+######################################################
+# If your audio files aren't accessible via a URL already, you can upload your audio file using this API
+upload_url= speect_to_text.upload_audio(file_location)
+
+# choose your service provider: "asemblyai", "aws", "whisper"
+service_provider="assembly"
+# transcribe
+transcribe_id= speect_to_text.transcribe(upload_url,service_provider)
+# get the status, the result will be out after the job is done, so we wait a bit :) 
+status=speect_to_text.status(transcribe_id,service_provider=service_provider)
+
+# get the text result
+## assembly: status["text"]
+## whisper: status["transcription_data"]["transcription"]
+## aws: retreive result from json file link
+
+print(status["text"])
+