Fix "Add online serving to Stable Audio Diffusion and introduce v1/audio/generate endpoint" (#1794)
Signed-off-by: Ekagra Ranjan <3116519+ekagra-ranjan@users.noreply.github.com> Signed-off-by: Yueqian Lin <linyueqian@outlook.com> Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Co-authored-by: Yueqian Lin <70319226+linyueqian@users.noreply.github.com> Co-authored-by: Yueqian Lin <linyueqian@outlook.com>
This commit is contained in:
13 files changed
+1788
-7
No files matched your search
@@ -124,6 +124,53 @@ class OpenAICreateSpeechRequest(BaseModel):
|
||||
return self
|
||||
|
||||
|
||||
class OpenAICreateAudioGenerateRequest(BaseModel):
|
||||
"""Request model for audio generation via diffusion models (e.g. Stable Audio)."""
|
||||
|
||||
input: str = Field(
|
||||
description="Text prompt describing the audio to generate",
|
||||
)
|
||||
model: str | None = None
|
||||
response_format: Literal["wav", "pcm", "flac", "mp3", "aac", "opus"] = "wav"
|
||||
speed: float | None = Field(
|
||||
default=1.0,
|
||||
ge=0.25,
|
||||
le=4.0,
|
||||
)
|
||||
stream_format: Literal["sse", "audio"] | None = "audio"
|
||||
audio_length: float | None = Field(
|
||||
default=None,
|
||||
description="Audio length in seconds",
|
||||
)
|
||||
audio_start: float | None = Field(
|
||||
default=0.0,
|
||||
description="Audio start time in seconds",
|
||||
)
|
||||
negative_prompt: str | None = Field(
|
||||
default=None,
|
||||
description="Negative prompt for classifier-free guidance",
|
||||
)
|
||||
guidance_scale: float | None = Field(
|
||||
default=None,
|
||||
description="Guidance scale for diffusion models",
|
||||
)
|
||||
num_inference_steps: int | None = Field(
|
||||
default=None,
|
||||
description="Number of inference steps",
|
||||
)
|
||||
seed: int | None = Field(
|
||||
default=None,
|
||||
description="Random seed for reproducibility",
|
||||
)
|
||||
|
||||
@field_validator("stream_format")
|
||||
@classmethod
|
||||
def validate_stream_format(cls, v: str) -> str:
|
||||
if v == "sse":
|
||||
raise ValueError("'sse' is not a supported stream_format yet. Please use 'audio'.")
|
||||
return v
|
||||
|
||||
|
||||
class CreateAudio(BaseModel):
|
||||
audio_tensor: np.ndarray
|
||||
sample_rate: int = 24000
|
||||
|
||||
Reference in new issue
Block a user