[Frontend] Add /v1/audio/translations OpenAI API endpoint (#19615)

Signed-off-by: Roger Wang <ywang@roblox.com>
Signed-off-by: NickLucche <nlucches@redhat.com>
Co-authored-by: Roger Wang <ywang@roblox.com>

vllm/entrypoints/openai/protocol.py

@@ -1947,3 +1947,190 @@ class TranscriptionResponseVerbose(OpenAIBaseModel):
 
     words: Optional[list[TranscriptionWord]] = None
     """Extracted words and their corresponding timestamps."""
+
+
+class TranslationResponseStreamChoice(OpenAIBaseModel):
+    delta: DeltaMessage
+    finish_reason: Optional[str] = None
+    stop_reason: Optional[Union[int, str]] = None
+
+
+class TranslationStreamResponse(OpenAIBaseModel):
+    id: str = Field(default_factory=lambda: f"trsl-{random_uuid()}")
+    object: Literal["translation.chunk"] = "translation.chunk"
+    created: int = Field(default_factory=lambda: int(time.time()))
+    model: str
+    choices: list[TranslationResponseStreamChoice]
+    usage: Optional[UsageInfo] = Field(default=None)
+
+
+class TranslationRequest(OpenAIBaseModel):
+    # Ordered by official OpenAI API documentation
+    # https://platform.openai.com/docs/api-reference/audio/createTranslation
+
+    file: UploadFile
+    """
+    The audio file object (not file name) to translate, in one of these
+    formats: flac, mp3, mp4, mpeg, mpga, m4a, ogg, wav, or webm.
+    """
+
+    model: Optional[str] = None
+    """ID of the model to use.
+    """
+
+    prompt: str = Field(default="")
+    """An optional text to guide the model's style or continue a previous audio
+    segment.
+
+    The [prompt](https://platform.openai.com/docs/guides/speech-to-text#prompting)
+    should match the audio language.
+    """
+
+    response_format: AudioResponseFormat = Field(default="json")
+    """
+    The format of the output, in one of these options: `json`, `text`, `srt`,
+    `verbose_json`, or `vtt`.
+    """
+
+    # TODO support additional sampling parameters
+    # --8<-- [start:translation-sampling-params]
+    temperature: float = Field(default=0.0)
+    """The sampling temperature, between 0 and 1.
+
+    Higher values like 0.8 will make the output more random, while lower values
+    like 0.2 will make it more focused / deterministic. If set to 0, the model
+    will use [log probability](https://en.wikipedia.org/wiki/Log_probability)
+    to automatically increase the temperature until certain thresholds are hit.
+    """
+    # --8<-- [end:translation-sampling-params]
+
+    # --8<-- [start:translation-extra-params]
+    language: Optional[str] = None
+    """The language of the input audio we translate from.
+
+    Supplying the input language in
+    [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) format
+    will improve accuracy.
+    """
+
+    stream: Optional[bool] = False
+    """Custom field not present in the original OpenAI definition. When set, 
+    it will enable output to be streamed in a similar fashion as the Chat
+    Completion endpoint. 
+    """
+    # Flattened stream option to simplify form data.
+    stream_include_usage: Optional[bool] = False
+    stream_continuous_usage_stats: Optional[bool] = False
+    # --8<-- [end:translation-extra-params]
+
+    # Default sampling parameters for translation requests.
+    _DEFAULT_SAMPLING_PARAMS: dict = {
+        "temperature": 0,
+    }
+
+    def to_sampling_params(
+            self,
+            default_max_tokens: int,
+            default_sampling_params: Optional[dict] = None) -> SamplingParams:
+        # TODO(#9845): remove max_tokens when field is removed from OpenAI API
+        max_tokens = default_max_tokens
+
+        if default_sampling_params is None:
+            default_sampling_params = {}
+        # Default parameters
+        if (temperature := self.temperature) is None:
+            temperature = default_sampling_params.get(
+                "temperature", self._DEFAULT_SAMPLING_PARAMS["temperature"])
+
+        return SamplingParams.from_optional(temperature=temperature,
+                                            max_tokens=max_tokens,
+                                            output_kind=RequestOutputKind.DELTA
+                                            if self.stream \
+                                            else RequestOutputKind.FINAL_ONLY)
+
+    @model_validator(mode="before")
+    @classmethod
+    def validate_stream_options(cls, data):
+        stream_opts = ["stream_include_usage", "stream_continuous_usage_stats"]
+        stream = data.get("stream", False)
+        if any(bool(data.get(so, False)) for so in stream_opts) and not stream:
+            raise ValueError(
+                "Stream options can only be defined when `stream=True`.")
+
+        return data
+
+
+# Translation response objects
+class TranslationResponse(OpenAIBaseModel):
+    text: str
+    """The translated text."""
+
+
+class TranslationWord(OpenAIBaseModel):
+    end: float
+    """End time of the word in seconds."""
+
+    start: float
+    """Start time of the word in seconds."""
+
+    word: str
+    """The text content of the word."""
+
+
+class TranslationSegment(OpenAIBaseModel):
+    id: int
+    """Unique identifier of the segment."""
+
+    avg_logprob: float
+    """Average logprob of the segment.
+
+    If the value is lower than -1, consider the logprobs failed.
+    """
+
+    compression_ratio: float
+    """Compression ratio of the segment.
+
+    If the value is greater than 2.4, consider the compression failed.
+    """
+
+    end: float
+    """End time of the segment in seconds."""
+
+    no_speech_prob: float
+    """Probability of no speech in the segment.
+
+    If the value is higher than 1.0 and the `avg_logprob` is below -1, consider
+    this segment silent.
+    """
+
+    seek: int
+    """Seek offset of the segment."""
+
+    start: float
+    """Start time of the segment in seconds."""
+
+    temperature: float
+    """Temperature parameter used for generating the segment."""
+
+    text: str
+    """Text content of the segment."""
+
+    tokens: list[int]
+    """Array of token IDs for the text content."""
+
+
+class TranslationResponseVerbose(OpenAIBaseModel):
+    duration: str
+    """The duration of the input audio."""
+
+    language: str
+    """The language of the input audio."""
+
+    text: str
+    """The translated text."""
+
+    segments: Optional[list[TranslationSegment]] = None
+    """Segments of the translated text and their corresponding details."""
+
+    words: Optional[list[TranslationWord]] = None
+    """Extracted words and their corresponding timestamps."""