dwani 0.1.13__py3-none-any.whl → 0.1.16__py3-none-any.whl

dwani/__init__.py

@@ -30,8 +30,8 @@ class chat:
 
 class audio:
     @staticmethod
-    def speech(input, response_format="wav"):
-        return _get_client().speech(input, response_format)
+    def speech(input, response_format="wav", language="kannada"):
+        return _get_client().speech(input, response_format, language)
 
 class vision:
     @staticmethod
@@ -53,8 +53,8 @@ class translate:
 
 class document:
     @staticmethod
-    def run_ocr(file_path, language="eng_Latn", model="gemma3"):
-        return _get_client().document_ocr(file_path, language, model)
+    def run_ocr(file_path, model="gemma3"):
+        return _get_client().document_ocr(file_path, model)
     
     @staticmethod
     def run_summarize(file_path, page_number=1, src_lang="eng_Latn", tgt_lang="kan_Knda", model="gemma3"):

dwani/audio.py

@@ -1,10 +1,11 @@
 from .exceptions import DwaniAPIError
 import requests
 
-def audio_speech(client, input, response_format="mp3", output_file=None):
+def audio_speech(client, input, response_format="mp3", output_file=None, language="kannada"):
     params = {
         "input": input,
-        "response_format": response_format
+        "response_format": response_format,
+        "language": language
     }
     resp = requests.post(
         f"{client.api_base}/v1/audio/speech",

dwani/client.py

@@ -27,9 +27,9 @@ class DwaniClient:
         from .chat import chat_direct
         return chat_direct(self, prompt=prompt, model=model, system_prompt=system_prompt)
 
-    def speech(self, input, response_format="mp3"):
+    def speech(self, input, response_format="wav", language="kannada"):
         from .audio import audio_speech
-        return audio_speech(self, input=input, response_format=response_format)
+        return audio_speech(self, input=input, response_format=response_format, language=language)
 
     def caption(self, file_path, query="describe the image", src_lang="eng_Latn", tgt_lang="kan_Knda", model="gemma3"):
         from .vision import vision_caption
@@ -43,9 +43,9 @@ class DwaniClient:
         from .asr import asr_transcribe
         return asr_transcribe(self, file_path=file_path, language=language)
     
-    def document_ocr(self, file_path, language=None, model="gemma3"):
+    def document_ocr(self, file_path, model="gemma3"):
         from .docs import document_ocr
-        return document_ocr(self, file_path=file_path, language=language, model=model)
+        return document_ocr(self, file_path=file_path, model=model)
 
     def document_summarize(self, file_path, page_number=1, src_lang="eng_Latn", tgt_lang="kan_Knda", model="gemma3"):
         from .docs import document_summarize

dwani/docs.py

@@ -40,21 +40,19 @@ def validate_model(model):
         raise ValueError(f"Unsupported model: {model}. Supported models: {VALID_MODELS}")
     return model
 
-def document_ocr(client, file_path, language=None, model="gemma3"):
+def document_ocr(client, file_path, model="gemma3"):
     """OCR a document (image/PDF) and return extracted text."""
-    logger.debug(f"Calling document_ocr: file_path={file_path}, language={language}, model={model}")
+    logger.debug(f"Calling document_ocr: file_path={file_path}, model={model}")
     validate_model(model)
     
     data = {"model": model}
-    if language:
-        data["language"] = normalize_language(language)
-    
+
     with open(file_path, "rb") as f:
         mime_type = "application/pdf" if file_path.lower().endswith('.pdf') else "image/png"
         files = {"file": (file_path, f, mime_type)}
         try:
             resp = requests.post(
-                f"{client.api_base}/v1/document/ocr",
+                f"{client.api_base}/v1/extract-text",
                 headers=client._headers(),
                 files=files,
                 data=data,
@@ -256,10 +254,10 @@ def doc_query_kannada(
 
 class Documents:
     @staticmethod
-    def ocr(file_path, language=None, model="gemma3"):
+    def run_ocr(file_path, model="gemma3"):
         from .client import DwaniClient
         client = DwaniClient()
-        return document_ocr(client, file_path, language, model)
+        return document_ocr(client, file_path, model)
     
     @staticmethod
     def summarize(file_path, page_number=1, src_lang="eng_Latn", tgt_lang="kan_Knda", model="gemma3"):

{dwani-0.1.13.dist-info → dwani-0.1.16.dist-info}/METADATA

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: dwani
-Version: 0.1.13
-Summary: Multimodal API for Indian languages (Chat, Vision, TTS, ASR, Translate, Docs)
+Version: 0.1.16
+Summary: Multimodal API for Indian + European languages (Chat, Vision, TTS, ASR, Translate, Docs)
 Author-email: sachin <python@dwani.ai>
 License: MIT License
         
@@ -32,10 +32,16 @@ Requires-Python: >=3.8
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: requests>=2.25.0
+Requires-Dist: openai
 Dynamic: license-file
 
 # dwani.ai - python library
 
+- dwani.ai is self-hosted GenAI platform for multimodal AI inference.
+
+- Image, Speech, Docs, Text supported today !!
+
+- dwani.ai - is now compatible with OpenAI Spec
 
 ### Install the library
 ```bash
@@ -43,7 +49,10 @@ pip install --upgrade dwani
 ```
 
 ### Languages supported
-    - Assamese, Bengali, Gujarati, Hindi, Kannada, Malayalam, Marathi, Odia, Punjabi, Tamil, Telugu
+- Indian
+  - Assamese, Bengali, Gujarati, Hindi, Kannada, Malayalam, Marathi Odia, Punjabi, Tamil, Telugu
+- European
+  - English, German
 
 ### Setup the credentials
 ```python
@@ -56,62 +65,56 @@ dwani.api_base = os.getenv("DWANI_API_BASE_URL")
 ```
 
 
-### Text Query 
 ---
-- With model selection
-  - gemma3 (default), qwen3, sarvam-m
 
----
-- gemma3 - with translation
-```python
-resp = dwani.Chat.create(prompt="Hello!", src_lang="english", tgt_lang="kannada", model="gemma3")
-print(resp)
-```
-```json
-{'response': 'ನಮಸ್ತೆ! ಭಾರತ ಮತ್ತು ಕರ್ನಾಟಕವನ್ನು ಗಮನದಲ್ಲಿಟ್ಟುಕೊಂಡು ಇಂದು ನಿಮ್ಮ ಪ್ರಶ್ನೆಗಳಿಗೆ ನಾನು ನಿಮಗೆ ಹೇಗೆ ಸಹಾಯ ಮಾಡಲಿ?'}
-```
+- Source Code : [https://github.com/dwani-ai/dwani-python-sdk](https://github.com/dwani-ai/dwani-python-sdk)
+- Check examples folder for detailed use cases 
 
-- gemma3 - without translation
+  - [examples/chat.py](examples/chat.py)
+  - [examples/vision.py](examples/vision.py)
+  - [examples/docs.py](examples/docs.py)
+  - [examples/speech.py](examples/speech.py)
+  - [examples/asr.py](examples/asr.py)
+
+#### Document - OCR
 ```python
-resp = dwani.Chat.direct(prompt="Hello!", model="gemma3")
-print(resp)
+result = dwani.Documents.run_ocr(file_path="dwani-workshop.pdf", model="gemma3")
+print(result)
 ```
 ```json
-{'response': 'Hello! I am Dwani, ready to assist you with information pertaining to India, specifically Karnataka. '}
+{'page_content': "Here's the plain text extracted from the image:\n\ndwani's Goals\n\nTo integrate and enhance the following models and services for Kannada:\n\n*   **Automatic Speech Recognition (ASR):**"}
 ```
 
+
+### Text Query 
+---
+- gemma3 (default)
+
+  ```python
+  resp = dwani.Chat.create(prompt="Hello!", src_lang="english", tgt_lang="kannada", model="gemma3")
+  print(resp)
+  ```
+  ```json
+  {'response': 'ನಮಸ್ತೆ! ಭಾರತ ಮತ್ತು ಕರ್ನಾಟಕವನ್ನು ಗಮನದಲ್ಲಿಟ್ಟುಕೊಂಡು ಇಂದು ನಿಮ್ಮ ಪ್ರಶ್ನೆಗಳಿಗೆ ನಾನು ನಿಮಗೆ ಹೇಗೆ ಸಹಾಯ ಮಾಡಲಿ?'}
+  ```
+
 ---
 ### Vision Query
 ---
-- With model selection
-  - gemma3 (default), moondream, smolvla
-
-- gemma3 - with translation
-```python
-result = dwani.Vision.caption(
-    file_path="image.png",
-    query="Describe this logo",
-    src_lang="english",
-    tgt_lang="kannada",
-    model="gemma3"
-)
-print(result)
-```
-```json
-{'answer': 'ಒಂದು ವಾಕ್ಯದಲ್ಲಿ ಚಿತ್ರದ ಸಾರಾಂಶವನ್ನು ಇಲ್ಲಿ ನೀಡಲಾಗಿದೆಃ ಪ್ರಕಟಣೆಯ ಅವಲೋಕನವು ಪ್ರಸ್ತುತ ಅರವತ್ತನಾಲ್ಕು ದೇಶಗಳು/ಪ್ರದೇಶಗಳನ್ನು ಸೇರಿಸಲಾಗಿದೆ ಮತ್ತು ಇನ್ನೂ ಹದಿನಾರು ಪ್ರದೇಶಗಳನ್ನು ಸೇರಿಸಬೇಕಾಗಿದೆ. ಒದಗಿಸಲಾದ ಚಿತ್ರದಲ್ಲಿ ಲಾಂಛನವು ಕಾಣಿಸುವುದಿಲ್ಲ.'}
-```
-- gemma3 - without translation
-```python
-result = dwani.Vision.caption_direct(
-    file_path="image.png",
-    query="Describe this logo",
-    model="gemma3"
-)
-print(result)
-```
-```json
- {'answer': 'The logo displays a publishing overview stating that changes are under review, with a production rollout initiated at version sixty-four point one point one, expanding to sixteen countries/regions including Australia and Bangladesh.'}
-```
+- gemma3 (default)
+    ```python
+    result = dwani.Vision.caption(
+        file_path="image.png",
+        query="Describe this logo",
+        src_lang="english",
+        tgt_lang="kannada",
+        model="gemma3"
+    )
+    print(result)
+    ```
+    ```json
+    {'answer': 'ಒಂದು ವಾಕ್ಯದಲ್ಲಿ ಚಿತ್ರದ ಸಾರಾಂಶವನ್ನು ಇಲ್ಲಿ ನೀಡಲಾಗಿದೆಃ ಪ್ರಕಟಣೆಯ ಅವಲೋಕನವು ಪ್ರಸ್ತುತ ಅರವತ್ತನಾಲ್ಕು ದೇಶಗಳು/ಪ್ರದೇಶಗಳನ್ನು ಸೇರಿಸಲಾಗಿದೆ ಮತ್ತು ಇನ್ನೂ ಹದಿನಾರು ಪ್ರದೇಶಗಳನ್ನು ಸೇರಿಸಬೇಕಾಗಿದೆ. ಒದಗಿಸಲಾದ ಚಿತ್ರದಲ್ಲಿ ಲಾಂಛನವು ಕಾಣಿಸುವುದಿಲ್ಲ.'}
+    ```
 
 ---
 ### Speech to Text -  Automatic Speech Recognition (ASR)
@@ -137,7 +140,7 @@ print(resp)
 ### Text to Speech -  Speech Synthesis
 ---
 ```python
-response = dwani.Audio.speech(input="ಕರ್ನಾಟಕದ ರಾಜಧಾನಿ ಯಾವುದು", response_format="wav")
+response = dwani.Audio.speech(input="ಕರ್ನಾಟಕದ ರಾಜಧಾನಿ ಯಾವುದು", response_format="wav", language="kannada")
 with open("output.wav", "wb") as f:
     f.write(response)
 ```

dwani-0.1.16.dist-info/RECORD

@@ -0,0 +1,14 @@
+dwani/__init__.py,sha256=2WxVFPYpwyZ68yqbzNOEpmwSsi7ksvw8-pZRmReUZCQ,3009
+dwani/asr.py,sha256=BAdqivQd57NJZX1dSY-J6EFi8TDdyuhf_AyCPcQ0M7w,1719
+dwani/audio.py,sha256=CFQrYU-KLwO7pCh_R7c1SSDJ6bugE5_av7lV8XTl-dY,936
+dwani/chat.py,sha256=Tui52XBhUyDyN2rOFoLme4oB0Q8fkD9_0tFDAnRzoaU,2979
+dwani/client.py,sha256=UvzmXShctntMmm1rIONVssv1c8HNgzBMZLOjxrCbp-4,3360
+dwani/docs.py,sha256=KSqmbVoImEFI_HK102iJwlemN3XQii2Mo7WOob2kFQE,10464
+dwani/exceptions.py,sha256=n06dPmR20rS4T3sJBWHQhGxzg4SJKzird9Hx0YTwwo0,226
+dwani/translate.py,sha256=c03N8-tN49IBcTA6GMOkrJ3MaVzZ12RnYdLQwRbEeoQ,2794
+dwani/vision.py,sha256=FviGewoV936CSv_K-latw0t3ZhSSCOF5LaGaq1oE4uA,3607
+dwani-0.1.16.dist-info/licenses/LICENSE,sha256=IAD8tbwWZbPWHXgYjabHoMv0aaUzZUYzYiEbfhTCisY,1070
+dwani-0.1.16.dist-info/METADATA,sha256=UKmC0j_PK9K8XcCqZZUuUbzPWbRi_bPlkO0BR6E9Xk8,6008
+dwani-0.1.16.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+dwani-0.1.16.dist-info/top_level.txt,sha256=AM5EhkyuO_EXQFR9JIxEV6tAYMCCyc-a1dLifpCGBUk,6
+dwani-0.1.16.dist-info/RECORD,,

dwani-0.1.13.dist-info/RECORD

@@ -1,14 +0,0 @@
-dwani/__init__.py,sha256=8Q1qdF0g6xWEy1_PK6qxG-EeZUp_QOwYBqNv_y6EyHI,3010
-dwani/asr.py,sha256=BAdqivQd57NJZX1dSY-J6EFi8TDdyuhf_AyCPcQ0M7w,1719
-dwani/audio.py,sha256=MWsIZazL91c2wa5AE1YY78l9RKaJwNFFHIajuwl43Jg,886
-dwani/chat.py,sha256=Tui52XBhUyDyN2rOFoLme4oB0Q8fkD9_0tFDAnRzoaU,2979
-dwani/client.py,sha256=Xqpc5tCOLpNWSPanY2eru2ywL6DytNCQjUDxv8hqXmw,3355
-dwani/docs.py,sha256=Cp0Gtudug79GH25toB-Npl35ZFA0TM32oZF2xH1VmNY,10598
-dwani/exceptions.py,sha256=n06dPmR20rS4T3sJBWHQhGxzg4SJKzird9Hx0YTwwo0,226
-dwani/translate.py,sha256=c03N8-tN49IBcTA6GMOkrJ3MaVzZ12RnYdLQwRbEeoQ,2794
-dwani/vision.py,sha256=FviGewoV936CSv_K-latw0t3ZhSSCOF5LaGaq1oE4uA,3607
-dwani-0.1.13.dist-info/licenses/LICENSE,sha256=IAD8tbwWZbPWHXgYjabHoMv0aaUzZUYzYiEbfhTCisY,1070
-dwani-0.1.13.dist-info/METADATA,sha256=PZSXwWoH6UzDF17X8_CfQIvzHC-Z1LjC3s3GwWeOUfc,5791
-dwani-0.1.13.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-dwani-0.1.13.dist-info/top_level.txt,sha256=AM5EhkyuO_EXQFR9JIxEV6tAYMCCyc-a1dLifpCGBUk,6
-dwani-0.1.13.dist-info/RECORD,,