dwani 0.1.3__tar.gz → 0.1.5__tar.gz

{dwani-0.1.3 → dwani-0.1.5}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: dwani
-Version: 0.1.3
-Summary: Multimodal AI server for Indian languages (speech, vision, LLMs, TTS, ASR, etc.)
+Version: 0.1.5
+Summary: Multimodal API for Indian languages (speech, vision, LLMs, TTS, ASR, etc.)
 Author-email: sachin <python@dwani.ai>
 License: MIT License
         
@@ -28,7 +28,7 @@ License: MIT License
 Project-URL: Homepage, https://github.com/dwani-ai/dwani-python
 Project-URL: Source, https://github.com/dwani-ai/dwani-python
 Project-URL: Issues, https://github.com/dwani-ai/dwani-python/issues
-Requires-Python: >=3.10
+Requires-Python: >=3.8
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: requests>=2.25.0
@@ -37,25 +37,64 @@ Dynamic: license-file
 # dwani.ai - python library
 
 
+### Install the library
 ```bash
 pip install dwani
 ```
 
-
-
+### Setup the credentials
 ```python
 import dwani
 import os
 
 dwani.api_key = os.getenv("DWANI_API_KEY")
 
-dwani.api_base = os.getenv("DWANI_API_BASE")
+dwani.api_base = os.getenv("DWANI_API_BASE_URL")
+```
 
-resp = dwani.Chat.create("Hello!", "eng_Latn", "kan_Knda")
+### Examples
+
+#### Text Query 
+```python
+resp = dwani.Chat.create(prompt="Hello!", src_lang="eng_Latn", tgt_lang="kan_Knda")
 print(resp)
 ```
 
+#### Vision Query
+```python
+result = dwani.Vision.caption(
+    file_path="image.png",
+    query="Describe this logo",
+    src_lang="eng_Latn",
+    tgt_lang="kan_Knda"
+)
+print(result)
+```
+
+#### Speech to Text -  Automatic Speech Recognition (ASR)
+```python
+result = dwani.ASR.transcribe(file_path="kannada_sample.wav", language="kannada")
+print(result)
+```
+
+#### Text to Speech -  Speech Synthesis
+
+```python
+response = dwani.Audio.speech(input="ಕರ್ನಾಟಕ ದ ರಾಜಧಾನಿ ಯಾವುದು", response_format="mp3")
+with open("output.mp3", "wb") as f:
+    f.write(response)
+```
+
+
+
+- Website -> [dwani.ai](https://dwani.ai)
+
+
 
+#### Contact
+- For any questions or issues, please open an issue on GitHub or contact us via email.
+- For collaborations
+  - Join the discord group - [invite link](https://discord.gg/WZMCerEZ2P) 
 <!-- 
 ## local development
 pip install -e .

dwani-0.1.5/README.md

@@ -0,0 +1,73 @@
+# dwani.ai - python library
+
+
+### Install the library
+```bash
+pip install dwani
+```
+
+### Setup the credentials
+```python
+import dwani
+import os
+
+dwani.api_key = os.getenv("DWANI_API_KEY")
+
+dwani.api_base = os.getenv("DWANI_API_BASE_URL")
+```
+
+### Examples
+
+#### Text Query 
+```python
+resp = dwani.Chat.create(prompt="Hello!", src_lang="eng_Latn", tgt_lang="kan_Knda")
+print(resp)
+```
+
+#### Vision Query
+```python
+result = dwani.Vision.caption(
+    file_path="image.png",
+    query="Describe this logo",
+    src_lang="eng_Latn",
+    tgt_lang="kan_Knda"
+)
+print(result)
+```
+
+#### Speech to Text -  Automatic Speech Recognition (ASR)
+```python
+result = dwani.ASR.transcribe(file_path="kannada_sample.wav", language="kannada")
+print(result)
+```
+
+#### Text to Speech -  Speech Synthesis
+
+```python
+response = dwani.Audio.speech(input="ಕರ್ನಾಟಕ ದ ರಾಜಧಾನಿ ಯಾವುದು", response_format="mp3")
+with open("output.mp3", "wb") as f:
+    f.write(response)
+```
+
+
+
+- Website -> [dwani.ai](https://dwani.ai)
+
+
+
+#### Contact
+- For any questions or issues, please open an issue on GitHub or contact us via email.
+- For collaborations
+  - Join the discord group - [invite link](https://discord.gg/WZMCerEZ2P) 
+<!-- 
+## local development
+pip install -e .
+
+
+pip install twine build
+rm -rf dist/
+python -m build
+
+python -m twine upload dist/*
+
+-->

dwani-0.1.3/dwani/vision.py → dwani-0.1.5/dwani/asr.py

@@ -1,21 +1,20 @@
 from .exceptions import DhwaniAPIError
 import requests
-def vision_caption(client, file_path, length="short"):
+def asr_transcribe(client, file_path, language):
     with open(file_path, "rb") as f:
         files = {"file": f}
-        data = {"length": length}
         resp = requests.post(
-            f"{client.api_base}/caption/",
+            f"{client.api_base}/v1/transcribe/?language={language}",
             headers=client._headers(),
-            files=files,
-            data=data
+            files=files
         )
     if resp.status_code != 200:
         raise DhwaniAPIError(resp)
     return resp.json()
 
-class Vision:
+class ASR:
     @staticmethod
-    def caption(*args, **kwargs):
+    def transcribe(*args, **kwargs):
         from . import _get_client
-        return _get_client().caption(*args, **kwargs)
+        return _get_client().transcribe(*args, **kwargs)
+

{dwani-0.1.3 → dwani-0.1.5}/dwani/audio.py

@@ -1,16 +1,16 @@
 from .exceptions import DhwaniAPIError
 import requests
-def audio_speech(client, input, voice, model, response_format="mp3", output_file=None):
-    data = {
+
+def audio_speech(client, input, response_format="mp3", output_file=None):
+    params = {
         "input": input,
-        "voice": voice,
-        "model": model,
         "response_format": response_format
     }
     resp = requests.post(
         f"{client.api_base}/v1/audio/speech",
-        headers={**client._headers(), "Content-Type": "application/json"},
-        json=data,
+        headers={**client._headers(), "accept": "application/json"},
+        params=params,
+        data='',  # Empty body, as in the curl example
         stream=True
     )
     if resp.status_code != 200:

{dwani-0.1.3 → dwani-0.1.5}/dwani/client.py

@@ -5,7 +5,7 @@ from .exceptions import DhwaniAPIError
 class DhwaniClient:
     def __init__(self, api_key=None, api_base=None):
         self.api_key = api_key or os.getenv("DWANI_API_KEY")
-        self.api_base = api_base or os.getenv("DWANI_API_BASE", "http://localhost:7860")
+        self.api_base = api_base or os.getenv("DWANI_API_BASE_URL", "http://localhost:8000")
         if not self.api_key:
             raise ValueError("DHWANI_API_KEY not set")
 
@@ -15,19 +15,23 @@ class DhwaniClient:
     def chat(self, prompt, src_lang, tgt_lang, **kwargs):
         from .chat import chat_create
         return chat_create(self, prompt, src_lang, tgt_lang, **kwargs)
-
+    
+    def translate(self, sentences, src_lang, tgt_lang, **kwargs):
+        from .translate import run_translate
+        return run_translate(self, sentences=sentences,src_lang= src_lang, tgt_lang=tgt_lang, **kwargs)
 
     def speech(self, *args, **kwargs):
         from .audio import audio_speech
         return audio_speech(self, *args, **kwargs)
 
-    def caption(self, *args, **kwargs):
+    def caption(self, file_path, query="describe the image", src_lang="eng_Latn", tgt_lang="kan_Knda"):
         from .vision import vision_caption
-        return vision_caption(self, *args, **kwargs)
+        return vision_caption(self, file_path, query, src_lang, tgt_lang)
 
     def transcribe(self, *args, **kwargs):
         from .asr import asr_transcribe
         return asr_transcribe(self, *args, **kwargs)
+    
     def document_ocr(self, file_path, language=None):
         from .docs import document_ocr
         return document_ocr(self, file_path, language)

dwani-0.1.5/dwani/translate.py

@@ -0,0 +1,29 @@
+from .exceptions import DhwaniAPIError
+import requests
+
+def run_translate(client, sentences, src_lang, tgt_lang, **kwargs):
+    url = f"{client.api_base}/v1/translate"
+    payload = {
+        "sentences": sentences,
+        "src_lang": src_lang,
+        "tgt_lang": tgt_lang
+    }
+    payload.update(kwargs)
+    resp = requests.post(
+        url,
+        headers={**client._headers(), "Content-Type": "application/json", "accept": "application/json"},
+        json=payload
+    )
+    if resp.status_code != 200:
+        raise DhwaniAPIError(resp)
+    return resp.json()
+
+class Translate:
+    @staticmethod
+    def translate(sentence, src_lang, tgt_lang, **kwargs):
+        from . import _get_client
+        client = _get_client()
+        # Ensure sentences is always a list
+        response = run_translate(client, [sentence], src_lang, tgt_lang, **kwargs)
+        # Return the first translation, or None if not found
+        return response.get("translations", [None])[0]

dwani-0.1.5/dwani/vision.py

@@ -0,0 +1,31 @@
+from .exceptions import DhwaniAPIError
+import requests
+def vision_caption(client, file_path, query="describe the image", src_lang="eng_Latn", tgt_lang="kan_Knda"):
+    # Build the endpoint using the client's api_base
+    url = (
+        f"{client.api_base}/v1/indic_visual_query"
+        f"?src_lang={src_lang}&tgt_lang={tgt_lang}"
+    )
+    headers = {
+        **client._headers(),
+        "accept": "application/json"
+        # Note: 'Content-Type' will be set automatically by requests when using 'files'
+    }
+    with open(file_path, "rb") as f:
+        files = {"file": (file_path, f, "image/png")}
+        data = {"query": query}
+        resp = requests.post(
+            url,
+            headers=headers,
+            files=files,
+            data=data
+        )
+    if resp.status_code != 200:
+        raise DhwaniAPIError(resp)
+    return resp.json()
+
+class Vision:
+    @staticmethod
+    def caption(*args, **kwargs):
+        from . import _get_client
+        return _get_client().caption(*args, **kwargs)

{dwani-0.1.3 → dwani-0.1.5}/dwani.egg-info/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: dwani
-Version: 0.1.3
-Summary: Multimodal AI server for Indian languages (speech, vision, LLMs, TTS, ASR, etc.)
+Version: 0.1.5
+Summary: Multimodal API for Indian languages (speech, vision, LLMs, TTS, ASR, etc.)
 Author-email: sachin <python@dwani.ai>
 License: MIT License
         
@@ -28,7 +28,7 @@ License: MIT License
 Project-URL: Homepage, https://github.com/dwani-ai/dwani-python
 Project-URL: Source, https://github.com/dwani-ai/dwani-python
 Project-URL: Issues, https://github.com/dwani-ai/dwani-python/issues
-Requires-Python: >=3.10
+Requires-Python: >=3.8
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: requests>=2.25.0
@@ -37,25 +37,64 @@ Dynamic: license-file
 # dwani.ai - python library
 
 
+### Install the library
 ```bash
 pip install dwani
 ```
 
-
-
+### Setup the credentials
 ```python
 import dwani
 import os
 
 dwani.api_key = os.getenv("DWANI_API_KEY")
 
-dwani.api_base = os.getenv("DWANI_API_BASE")
+dwani.api_base = os.getenv("DWANI_API_BASE_URL")
+```
 
-resp = dwani.Chat.create("Hello!", "eng_Latn", "kan_Knda")
+### Examples
+
+#### Text Query 
+```python
+resp = dwani.Chat.create(prompt="Hello!", src_lang="eng_Latn", tgt_lang="kan_Knda")
 print(resp)
 ```
 
+#### Vision Query
+```python
+result = dwani.Vision.caption(
+    file_path="image.png",
+    query="Describe this logo",
+    src_lang="eng_Latn",
+    tgt_lang="kan_Knda"
+)
+print(result)
+```
+
+#### Speech to Text -  Automatic Speech Recognition (ASR)
+```python
+result = dwani.ASR.transcribe(file_path="kannada_sample.wav", language="kannada")
+print(result)
+```
+
+#### Text to Speech -  Speech Synthesis
+
+```python
+response = dwani.Audio.speech(input="ಕರ್ನಾಟಕ ದ ರಾಜಧಾನಿ ಯಾವುದು", response_format="mp3")
+with open("output.mp3", "wb") as f:
+    f.write(response)
+```
+
+
+
+- Website -> [dwani.ai](https://dwani.ai)
+
+
 
+#### Contact
+- For any questions or issues, please open an issue on GitHub or contact us via email.
+- For collaborations
+  - Join the discord group - [invite link](https://discord.gg/WZMCerEZ2P) 
 <!-- 
 ## local development
 pip install -e .

{dwani-0.1.3 → dwani-0.1.5}/dwani.egg-info/SOURCES.txt

@@ -8,6 +8,7 @@ dwani/chat.py
 dwani/client.py
 dwani/docs.py
 dwani/exceptions.py
+dwani/translate.py
 dwani/vision.py
 dwani.egg-info/PKG-INFO
 dwani.egg-info/SOURCES.txt

{dwani-0.1.3 → dwani-0.1.5}/pyproject.toml

@@ -4,14 +4,14 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "dwani"
-version = "0.1.3"
-description = "Multimodal AI server for Indian languages (speech, vision, LLMs, TTS, ASR, etc.)"
+version = "0.1.5"
+description = "Multimodal API for Indian languages (speech, vision, LLMs, TTS, ASR, etc.)"
 authors = [
     { name="sachin", email="python@dwani.ai" }
 ]
 readme = "README.md"
 license = { file = "LICENSE" }
-requires-python = ">=3.10"
+requires-python = ">=3.8"
 
 dependencies = [
     "requests>=2.25.0",

dwani-0.1.3/README.md

@@ -1,34 +0,0 @@
-# dwani.ai - python library
-
-
-```bash
-pip install dwani
-```
-
-
-
-```python
-import dwani
-import os
-
-dwani.api_key = os.getenv("DWANI_API_KEY")
-
-dwani.api_base = os.getenv("DWANI_API_BASE")
-
-resp = dwani.Chat.create("Hello!", "eng_Latn", "kan_Knda")
-print(resp)
-```
-
-
-<!-- 
-## local development
-pip install -e .
-
-
-pip install twine build
-rm -rf dist/
-python -m build
-
-python -m twine upload dist/*
-
--->

dwani-0.1.3/dwani/asr.py

@@ -1,37 +0,0 @@
-from .exceptions import DhwaniAPIError
-import requests
-def asr_transcribe(client, file_path, language):
-    with open(file_path, "rb") as f:
-        files = {"file": f}
-        resp = requests.post(
-            f"{client.api_base}/transcribe/?language={language}",
-            headers=client._headers(),
-            files=files
-        )
-    if resp.status_code != 200:
-        raise DhwaniAPIError(resp)
-    return resp.json()
-
-class ASR:
-    @staticmethod
-    def transcribe(*args, **kwargs):
-        from . import _get_client
-        return _get_client().transcribe(*args, **kwargs)
-
-
-'''
-from .docs import Documents
-
-class documents:
-    @staticmethod
-    def ocr(file_path, language=None):
-        return _get_client().document_ocr(file_path, language)
-
-    @staticmethod
-    def translate(file_path, src_lang, tgt_lang):
-        return _get_client().document_translate(file_path, src_lang, tgt_lang)
-
-    @staticmethod
-    def summarize(file_path, language=None):
-        return _get_client().document_summarize(file_path, language)
-'''