Update sdk

README.md

@@ -1,3 +1,74 @@
-# 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. Do a transcription job with three lines of code
+```
+import speechtotext
+
+speect_to_text = speechtotext.SpeechToText()
+transcribe_id= speect_to_text.transcribe(audio_url,service_provider="aws")
+status=speect_to_text.status(transcribe_id,service_provider=service_provider)
+
+```
+There is a daily limit (throttling) on the number of calls the user performs, transcription jobs can be done without specifying a token (API key). If the user intends on using the service more frequently, it is recommended to generate an token or API key from the dashboard @ [Slashml.com](https://www.slashml.com/).
+
+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]
+```
+or including it in your code as follows:
+```
+import speechtotext
+API_KEY="your_api_key"
+speect_to_text = speechtotext.SpeechToText(API_KEY)
+transcribe_id= speect_to_text.transcribe(audio_url,service_provider="aws")
+status=speect_to_text.status(transcribe_id,service_provider=service_provider)
+
+```
+
+-- 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

src/requirements.txt

@@ -0,0 +1 @@
+requests==2.28.1

src/speechtotext.py

@@ -0,0 +1,75 @@
+import requests
+import os
+from pathlib import Path
+import json
+
+class SpeechToText:
+
+    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/'
+    SLASHML_STATUS_URL = SLASHML_BASE_URL+'/transcription'
+    SLASHML_TRANSCRIPT_STATUS_URL = lambda self,id: f"{SpeechToText.SLASHML_STATUS_URL}/{id}/"
+
+    HEADERS:dict = {}
+    ## add the api key to sys path envs 
+    def __init__(self,API_KEY: str = None):
+        self.API_KEY=None
+        if API_KEY:
+            token="Token {0}".format(API_KEY)
+            self.HEADERS = {'authorization': token}
+           # print("Auth with "+API_KEY+"\nMake sure this matches your API Key generated in the dashboard settings")
+        elif os.environ.get('SLASHML_API_KEY'):
+            key_env =  os.environ.get('SLASHML_API_KEY')
+            token="Token {0}".format(key_env)
+            self.HEADERS = {'authorization': token}
+           # print("Auth with environment variable SLASHML_API_KEY")
+        else:
+            self.HEADERS=None 
+            print("No Auth, there are certain limites to the usage")
+
+    def upload_audio(self, file_location:str,header=None):
+        headers = self.HEADERS
+        files=[
+        ('audio',('test_audio.mp3',open(file_location,'rb'),'audio/mpeg'))
+        ]
+        response = requests.post(self.SLASHML_UPLOAD_URL,
+                                headers=headers,files=files)
+        return  response.json()["upload_url"]
+
+    def transcribe(self,upload_url:str,  service_provider: str,header=None ):
+
+        transcript_request = {'audio_url': upload_url}
+        headers = self.HEADERS
+        payload = {
+        "uploaded_audio_url": upload_url,
+        "service_provider": service_provider
+        }
+        response = requests.post(self.SLASHML_TRANSCRIPT_URL, headers=headers, data=payload)
+        return response.json()["id"]
+
+    def status(self,job_id:str, service_provider: str ,header=None):
+        headers = self.HEADERS
+        payload = {
+        "service_provider": service_provider
+        }
+
+        response = requests.get(self.SLASHML_TRANSCRIPT_STATUS_URL(job_id) , headers=headers, data=payload)
+
+        # Check the status code of the response
+        if response.status_code == 200:
+            return response.json()["transcription_data"]["transcription"]
+
+        elif response.status_code == 404:
+            return "NOT FOUND"
+
+        elif response.status_code == 500:
+            return "SERVER ERROR"
+        else:
+
+            return "UNKNOWN ERROR"
+
+    #    return response.json()["transcription_data"]["transcription"]
+
+
+    #### chgeck the code of the reponse, ifit is not 200, print back the error.

src/status_test.py

@@ -0,0 +1,24 @@
+import speechtotext
+import os 
+
+#API KEY (optional)
+#API_KEY="0d91bfede9c5c9de6ff1d5610ef71c3b6d5be9ee"
+# set environment path
+os.environ["SLASHML_API_KEY"] = "0d91bfede9c5c9de6ff1d5610ef71c3b6d5be9ee"
+# Initialize SlashML 
+speect_to_text = speechtotext.SpeechToText()
+# optional local file to upload, if not an accessible url
+# file_location="/Users/JJneid/Desktop/SlashMl/s2t_experiments/api_tests/test_french_english.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)
+# print(upload_url)
+# choose your service provider: "asembly", "aws", "whisper"
+service_provider="aws"
+# transcribe
+transcribe_id= "e5f43d08-6fa4-4cd7-86f9-2fb172004032"
+
+# 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: status["transcription_data"]["transcription"]
+print(status)
+#print(status["transcription_data"]["transcription"])

src/test_speechtotext_1.py

@@ -0,0 +1,21 @@
+import speechtotext
+import os 
+
+#API KEY (optional)
+
+# set environment path
+os.environ["SLASHML_API_KEY"] = "0d91bfede9c5c9de6ff1d5610ef71c3b6d5be9ee"
+# Initialize SlashML 
+speect_to_text = speechtotext.SpeechToText()
+# optional local file to upload, if not an accessible url
+file_location="/Users/JJneid/Desktop/SlashMl/s2t_experiments/api_tests/2test_audio.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: "asembly", "aws", "whisper"
+service_provider="aws"
+# transcribe
+transcribe_id= speect_to_text.transcribe(upload_url,service_provider)
+print(transcribe_id)
+
+status=speect_to_text.status(transcribe_id,service_provider=service_provider)
+print(status)