smallestai 4.0.1__tar.gz → 4.1.0__tar.gz

{smallestai-4.0.1/smallestai.egg-info → smallestai-4.1.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: smallestai
-Version: 4.0.1
+Version: 4.1.0
 Summary: Official Python client for the Smallest AI API
 Author-email: Smallest <support@smallest.ai>
 License: MIT

{smallestai-4.0.1 → smallestai-4.1.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "smallestai"
-version = "4.0.1"
+version = "4.1.0"
 description = "Official Python client for the Smallest AI API"
 authors = [
     {name = "Smallest", email = "support@smallest.ai"},

{smallestai-4.0.1 → smallestai-4.1.0}/smallestai/waves/async_waves_client.py

@@ -6,8 +6,8 @@ import aiofiles
 import requests
 from typing import Optional, Union, List
 
-from smallestai.waves.exceptions import TTSError, APIError
-from smallestai.waves.utils import (TTSOptions, validate_input,
+from smallestai.waves.exceptions import InvalidError, APIError
+from smallestai.waves.utils import (TTSOptions, validate_input, validate_asr_input,
                      get_smallest_languages, get_smallest_models, ALLOWED_AUDIO_EXTENSIONS, API_BASE_URL)
 
 
@@ -52,7 +52,7 @@ class AsyncWavesClient:
         """
         self.api_key = api_key or os.environ.get("SMALLEST_API_KEY")
         if not self.api_key:
-            raise TTSError()
+            raise InvalidError()
         if model == "lightning-large" and voice_id is None:
             voice_id = "lakshya"
 
@@ -150,7 +150,7 @@ class AsyncWavesClient:
             - Otherwise, returns the synthesized audio content as bytes.
 
         Raises:
-        - TTSError: If the provided file name does not have a .wav or .mp3 extension when `save_as` is specified.
+        - InvalidError: If the provided file name does not have a .wav or .mp3 extension when `save_as` is specified.
         - APIError: If the API request fails or returns an error.
         - ValueError: If an unexpected parameter is passed in `kwargs`.
         """
@@ -223,17 +223,17 @@ class AsyncWavesClient:
         - str: The response from the API as a formatted JSON string.
 
         Raises:
-        - TTSError: If the file does not exist or is not a valid audio file.
+        - InvalidError: If the file does not exist or is not a valid audio file.
         - APIError: If the API request fails or returns an error.
         """
         url = f"{API_BASE_URL}/lightning-large/add_voice"
 
         if not os.path.exists(file_path):
-            raise TTSError("Invalid file path. File does not exist.")
+            raise InvalidError("Invalid file path. File does not exist.")
 
         file_extension = os.path.splitext(file_path)[1].lower()
         if file_extension not in ALLOWED_AUDIO_EXTENSIONS:
-            raise TTSError(f"Invalid file type. Supported formats are: {ALLOWED_AUDIO_EXTENSIONS}")
+            raise InvalidError(f"Invalid file type. Supported formats are: {ALLOWED_AUDIO_EXTENSIONS}")
 
         headers = {
             'Authorization': f"Bearer {self.api_key}",
@@ -255,7 +255,7 @@ class AsyncWavesClient:
                 if res.status != 200:
                     raise APIError(f"Failed to add voice: {await res.text()}. For more information, visit https://waves.smallest.ai/")
 
-                return json.dumps(await res.json(), indent=4, ensure_ascii=False)
+                return await res.json()
         
         finally:
             if should_cleanup and self.session:
@@ -290,8 +290,61 @@ class AsyncWavesClient:
                 if res.status != 200:
                     raise APIError(f"Failed to delete voice: {await res.text()}. For more information, visit https://waves.smallest.ai/")
 
-                return json.dumps(await res.json(), indent=4, ensure_ascii=False)
+                return await res.json()
         finally:
             if should_cleanup and self.session:
                 await self.session.close()
-                self.session = None
+                self.session = None
+                
+    async def transcribe(
+        self,
+        file_path: str,
+        language: Optional[str] = "en",
+        word_timestamps: Optional[bool] = False,
+        age_detection: Optional[bool] = False,
+        gender_detection: Optional[bool] = False,
+        emotion_detection: Optional[bool] = False,
+        model: Optional[str] = "lightning"
+    ) -> dict:
+        validate_asr_input(file_path, model, language)
+
+        url = f"{API_BASE_URL}/speech-to-text"
+        headers = {
+            'Authorization': f"Bearer {self.api_key}",
+        }
+
+        should_cleanup = await self._ensure_session()
+
+        try:
+            file_extension = os.path.splitext(file_path)[1].lower()
+            content_type = f"audio/{file_extension[1:]}" if file_extension else "application/octet-stream"
+
+            async with aiofiles.open(file_path, 'rb') as f:
+                file_data = await f.read()
+
+            form = aiohttp.FormData()
+            form.add_field(
+                'file',
+                file_data,
+                filename=os.path.basename(file_path),
+                content_type=content_type
+            )
+            # Send options as multipart form fields (not query params)
+            form.add_field('model', model)
+            form.add_field('language', language)
+            form.add_field('word_timestamps', str(bool(word_timestamps)).lower())
+            form.add_field('age_detection', str(bool(age_detection)).lower())
+            form.add_field('gender_detection', str(bool(gender_detection)).lower())
+            form.add_field('emotion_detection', str(bool(emotion_detection)).lower())
+
+            async with self.session.post(url, headers=headers, data=form) as res:
+                if res.status != 200:
+                    raise APIError(
+                        f"Failed to transcribe audio: {await res.text()}. "
+                        "For more information, visit https://waves-docs.smallest.ai/v4.0.0/content/api-references/asr-post-api"
+                    )
+                return await res.json()
+        finally:
+            if should_cleanup and self.session:
+                await self.session.close()
+                self.session = None

{smallestai-4.0.1 → smallestai-4.1.0}/smallestai/waves/exceptions.py

@@ -1,18 +1,18 @@
-class TTSError(Exception):
+class InvalidError(Exception):
     """Base exception for TTS SDK"""
     default_message = "API key is required. Please set the `SMALLEST_API_KEY` environment variable or visit https://waves.smallest.ai/ to obtain your API key."
     
     def __init__(self, message=None):
         super().__init__(message or self.default_message)
 
-class APIError(TTSError):
+class APIError(InvalidError):
     """Raised when the API returns an error"""
     pass
 
-class ValidationError(TTSError):
+class ValidationError(InvalidError):
     """Raised when input validation fails"""
     pass
 
-class AuthenticationError(TTSError):
+class AuthenticationError(InvalidError):
     """Raised when authentication fails"""
     pass

{smallestai-4.0.1 → smallestai-4.1.0}/smallestai/waves/models.py

@@ -6,3 +6,11 @@ TTSModels = [
     "lightning-large",
     "lightning-v2"
 ]
+ASRLanguages_lightning = [
+    "it", "es", "en", "pt", "hi", "de", "fr", "uk", "ru", "kn", "ml", "pl",
+    "mr", "gu", "cs", "sk", "te", "or", "nl", "bn", "lv", "et", "ro", "pa",
+    "fi", "sv", "bg", "ta", "hu", "da", "lt", "mt", "multi"
+]
+ASRModels = [
+    "lightning"
+]

{smallestai-4.0.1 → smallestai-4.1.0}/smallestai/waves/utils.py

@@ -1,9 +1,10 @@
+import os
 from typing import List
 from typing import Optional
 from dataclasses import dataclass
 
 from smallestai.waves.exceptions import ValidationError
-from smallestai.waves.models import TTSModels, TTSLanguages_lightning, TTSLanguages_lightning_large, TTSLanguages_lightning_v2
+from smallestai.waves.models import TTSModels, TTSLanguages_lightning, TTSLanguages_lightning_large, TTSLanguages_lightning_v2, ASRModels, ASRLanguages_lightning
 
 
 API_BASE_URL = "https://waves-api.smallest.ai/api/v1"
@@ -25,7 +26,24 @@ class TTSOptions:
     enhancement: int
     language: str
     output_format: str
+    
+@dataclass
+class ASROptions:
+    model: str
+    api_key: str
+    language: str
+    word_timestamps: bool
+    age_detection: bool
+    gender_detection: bool
+    emotion_detection: bool
 
+def validate_asr_input(file_path: str, model: str, language: str):
+    if not os.path.isfile(file_path):
+        raise ValidationError("Invalid file path. File does not exist.")
+    if model not in ASRModels:
+        raise ValidationError(f"Invalid model: {model}. Must be one of {ASRModels}")
+    if language not in ASRLanguages_lightning:
+        raise ValidationError(f"Invalid language: {language}. Must be one of {ASRLanguages_lightning}")
 
 def validate_input(text: str, model: str, sample_rate: int, speed: float, consistency: Optional[float] = None, similarity: Optional[float] = None, enhancement: Optional[int] = None):
     if not text:

{smallestai-4.0.1 → smallestai-4.1.0}/smallestai/waves/waves_client.py

@@ -4,8 +4,8 @@ import copy
 import requests
 from typing import Optional, Union, List
 
-from smallestai.waves.exceptions import TTSError, APIError
-from smallestai.waves.utils import (TTSOptions, validate_input, 
+from smallestai.waves.exceptions import InvalidError, APIError
+from smallestai.waves.utils import (TTSOptions, validate_input, validate_asr_input,
                         get_smallest_languages, get_smallest_models, ALLOWED_AUDIO_EXTENSIONS, API_BASE_URL)
 
 class WavesClient:
@@ -48,7 +48,7 @@ class WavesClient:
         """
         self.api_key = api_key or os.environ.get("SMALLEST_API_KEY")
         if not self.api_key:
-            raise TTSError()
+            raise InvalidError()
         if model == "lightning-large" and voice_id is None:
             voice_id = "lakshya"
 
@@ -125,7 +125,7 @@ class WavesClient:
             - Otherwise, returns the synthesized audio content as bytes.
 
         Raises:
-        - TTSError: If the provided file name does not have a .wav or .mp3 extension when `save_as` is specified.
+        - InvalidError: If the provided file name does not have a .wav or .mp3 extension when `save_as` is specified.
         - APIError: If the API request fails or returns an error.
         """
         opts = copy.deepcopy(self.opts)
@@ -184,15 +184,15 @@ class WavesClient:
         - str: The response from the API as a formatted JSON string.
 
         Raises:
-        - TTSError: If the file does not exist or is not a valid audio file.
+        - InvalidError: If the file does not exist or is not a valid audio file.
         - APIError: If the API request fails or returns an error.
         """
         if not os.path.isfile(file_path):
-            raise TTSError("Invalid file path. File does not exist.")
+            raise InvalidError("Invalid file path. File does not exist.")
         
         file_extension = os.path.splitext(file_path)[1].lower()
         if file_extension not in ALLOWED_AUDIO_EXTENSIONS:
-            raise TTSError(f"Invalid file type. Supported formats are: {ALLOWED_AUDIO_EXTENSIONS}")
+            raise InvalidError(f"Invalid file type. Supported formats are: {ALLOWED_AUDIO_EXTENSIONS}")
         
         url = f"{API_BASE_URL}/lightning-large/add_voice"
         payload = {'displayName': display_name}
@@ -206,8 +206,8 @@ class WavesClient:
         response = requests.post(url, headers=headers, data=payload, files=files)
         if response.status_code != 200:
             raise APIError(f"Failed to add voice: {response.text}. For more information, visit https://waves.smallest.ai/")
-        
-        return json.dumps(response.json(), indent=4, ensure_ascii=False)
+
+        return response.json()
 
 
     def delete_voice(self, voice_id: str) -> str:
@@ -234,4 +234,41 @@ class WavesClient:
         if response.status_code != 200:
             raise APIError(f"Failed to delete voice: {response.text}. For more information, visit https://waves.smallest.ai/")
         
-        return json.dumps(response.json(), indent=4, ensure_ascii=False)
+        return response.json()
+    
+    def transcribe(
+        self,
+        file_path: str,
+        language: Optional[str] = "en",
+        word_timestamps: Optional[bool] = False,
+        age_detection: Optional[bool] = False,
+        gender_detection: Optional[bool] = False,
+        emotion_detection: Optional[bool] = False,
+        model: Optional[str] = "lightning"
+    ) -> dict:
+        validate_asr_input(file_path, model, language)
+
+        url = f"{API_BASE_URL}/speech-to-text"
+        headers = {
+            'Authorization': f"Bearer {self.api_key}",
+        }
+        payload = {
+            'model': model,
+            'language': language,
+            'word_timestamps': str(bool(word_timestamps)).lower(),
+            'age_detection': str(bool(age_detection)).lower(),
+            'gender_detection': str(bool(gender_detection)).lower(),
+            'emotion_detection': str(bool(emotion_detection)).lower()
+        }
+
+        file_extension = os.path.splitext(file_path)[1].lower()
+        content_type = f"audio/{file_extension[1:]}" if file_extension else "application/octet-stream"
+
+        with open(file_path, 'rb') as f:
+            files = {'file': (os.path.basename(file_path), f, content_type)}
+            response = requests.post(url, headers=headers, files=files, data=payload)
+
+        if response.status_code != 200:
+            raise APIError(f"Failed to transcribe audio: {response.text}. For more information, visit https://waves-docs.smallest.ai/v4.0.0/content/api-references/asr-post-api")
+
+        return response.json()

{smallestai-4.0.1 → smallestai-4.1.0/smallestai.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: smallestai
-Version: 4.0.1
+Version: 4.1.0
 Summary: Official Python client for the Smallest AI API
 Author-email: Smallest <support@smallest.ai>
 License: MIT