feat(audio): add speech, transcriptions and translations

README.md

@@ -39,6 +39,7 @@ print(response.choices[0].message.content)
 - **Embeddings**: Text embedding generation
 - **Vision/Image**: Analyze images with multimodal models
 - **Audio I/O**: Audio input and output support
+- **Audio API**: Text to speech, transcription and translation (`client.audio`)
 - **Video**: Video content analysis
 - **Web Search**: Real-time web search integration
 - **Context Compression**: Compress long conversation history to save tokens
@@ -154,6 +155,31 @@ response = client.chat.completions.create(
 print(response.choices[0].message.audio.transcript)
 ```
 
+### Speech, Transcription and Translation
+
+```python
+# Text to speech
+speech = client.audio.speech.create(
+    model="gpt-4o-mini-tts",
+    input="Hello from Apertis!",
+    voice="alloy",
+)
+speech.write_to_file("hello.mp3")
+
+# Speech to text: pass a path, an open file, bytes, or (filename, bytes)
+transcript = client.audio.transcriptions.create(file="meeting.mp3", model="whisper-1")
+print(transcript.text)
+
+# Subtitles come back as plain text
+srt = client.audio.transcriptions.create(
+    file="meeting.mp3", model="whisper-1", response_format="srt"
+)
+
+# Translate speech into English text
+translation = client.audio.translations.create(file="bonjour.m4a", model="whisper-1")
+print(translation.text)
+```
+
 ### Web Search
 
 Enable real-time web search for a non-free model by appending `:web` to its model ID.
@@ -581,6 +607,8 @@ the `publish` confirmation value.
 ## Changelog
 
 ### Unreleased
+- Add `client.audio`: `speech.create()` (binary audio with `write_to_file()`),
+  `transcriptions.create()` and `translations.create()` (multipart uploads).
 - Add `stream`, `thinking` and `extra_body` to `messages.create()`, with typed stream
   events and `thinking` / `redacted_thinking` content blocks.
 - Streamed requests that fail with an HTTP error now raise `APIError` (previously

apertis/_base_client.py

@@ -71,10 +71,20 @@ def request(
         json: Optional[Dict[str, Any]] = None,
         headers: Optional[Mapping[str, str]] = None,
         stream: bool = False,
+        data: Optional[Dict[str, Any]] = None,
+        files: Optional[Dict[str, Any]] = None,
     ) -> httpx.Response:
-        """Make an HTTP request."""
+        """Make an HTTP request.
+
+        With `files`, the body is multipart/form-data built from `data` and `files`;
+        file contents must already be bytes so a retry resends them unchanged.
+        """
         url = self._build_url(path)
         request_headers = self._build_headers(headers)
+        if files is not None:
+            # httpx sets the multipart content type with its boundary.
+            for key in [k for k in request_headers if k.lower() == "content-type"]:
+                del request_headers[key]
 
         retries = 0
         last_exception: Optional[Exception] = None
@@ -85,6 +95,8 @@ def request(
                     method,
                     url,
                     json=json,
+                    data=data,
+                    files=files,
                     headers=request_headers,
                     extensions={"stream": stream} if stream else None,
                 )
@@ -196,10 +208,20 @@ async def request(
         json: Optional[Dict[str, Any]] = None,
         headers: Optional[Mapping[str, str]] = None,
         stream: bool = False,
+        data: Optional[Dict[str, Any]] = None,
+        files: Optional[Dict[str, Any]] = None,
     ) -> httpx.Response:
-        """Make an HTTP request."""
+        """Make an HTTP request.
+
+        With `files`, the body is multipart/form-data built from `data` and `files`;
+        file contents must already be bytes so a retry resends them unchanged.
+        """
         url = self._build_url(path)
         request_headers = self._build_headers(headers)
+        if files is not None:
+            # httpx sets the multipart content type with its boundary.
+            for key in [k for k in request_headers if k.lower() == "content-type"]:
+                del request_headers[key]
 
         retries = 0
         last_exception: Optional[Exception] = None
@@ -210,6 +232,8 @@ async def request(
                     method,
                     url,
                     json=json,
+                    data=data,
+                    files=files,
                     headers=request_headers,
                     extensions={"stream": stream} if stream else None,
                 )

apertis/_client.py

@@ -5,6 +5,7 @@
 from typing import Mapping
 
 from apertis._base_client import AsyncClient, SyncClient
+from apertis.resources.audio import AsyncAudio, Audio
 from apertis.resources.chat import AsyncChat, Chat
 from apertis.resources.embeddings import AsyncEmbeddings, Embeddings
 from apertis.resources.models import AsyncModels, Models
@@ -58,6 +59,7 @@ def __init__(
         self.responses = Responses(self._client)
         self.messages = Messages(self._client)
         self.rerank = Rerank(self._client)
+        self.audio = Audio(self._client)
 
     def close(self) -> None:
         """Close the underlying HTTP client."""
@@ -115,6 +117,7 @@ def __init__(
         self.responses = AsyncResponses(self._client)
         self.messages = AsyncMessages(self._client)
         self.rerank = AsyncRerank(self._client)
+        self.audio = AsyncAudio(self._client)
 
     async def close(self) -> None:
         """Close the underlying HTTP client."""