public-mirrors/Kokoro-FastAPI
1
0
Fork 0
mirror of https://github.com/remsky/Kokoro-FastAPI.git synced 2025-08-05 16:48:53 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

fix: download_format option for audio response, handling in create_speech

Browse source
This commit is contained in:
remsky 2025-02-13 00:04:21 -07:00
parent 127aae4fab
commit 37ea01eaf9
3 changed files with 10 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
api/src/routers/openai_compatible.py
View file

@ -197,7 +197,9 @@ async def create_speech(
if request.return_download_link: if request.return_download_link:
from ..services.temp_manager import TempFileWriter from ..services.temp_manager import TempFileWriter
temp_writer = TempFileWriter(request.response_format) # Use download_format if specified, otherwise use response_format
output_format = request.download_format or request.response_format
temp_writer = TempFileWriter(output_format)
await temp_writer.__aenter__() # Initialize temp file await temp_writer.__aenter__() # Initialize temp file
# Get download path immediately after temp file creation # Get download path immediately after temp file creation
@ -205,7 +207,7 @@ async def create_speech(
# Create response headers with download path # Create response headers with download path
headers = { headers = {
"Content-Disposition": f"attachment; filename=speech.{request.response_format}", "Content-Disposition": f"attachment; filename=speech.{output_format}",
"X-Accel-Buffering": "no", "X-Accel-Buffering": "no",
"Cache-Control": "no-cache", "Cache-Control": "no-cache",
"Transfer-Encoding": "chunked", "Transfer-Encoding": "chunked",

4
api/src/structures/schemas.py
View file

@ -60,6 +60,10 @@ class OpenAISpeechRequest(BaseModel):
default="mp3", default="mp3",
description="The format to return audio in. Supported formats: mp3, opus, flac, wav, pcm. PCM format returns raw 16-bit samples without headers. AAC is not currently supported.", description="The format to return audio in. Supported formats: mp3, opus, flac, wav, pcm. PCM format returns raw 16-bit samples without headers. AAC is not currently supported.",
) )
download_format: Optional[Literal["mp3", "opus", "aac", "flac", "wav", "pcm"]] = Field(
default=None,
description="Optional different format for the final download. If not provided, uses response_format.",
)
speed: float = Field( speed: float = Field(
default=1.0, default=1.0,
ge=0.25, ge=0.25,

3
web/src/services/AudioService.js
View file

@ -39,7 +39,8 @@ export class AudioService {
body: JSON.stringify({ body: JSON.stringify({
input: text, input: text,
voice: voice, voice: voice,
response_format: 'mp3', response_format: 'mp3', // Always use mp3 for streaming playback
download_format: document.getElementById('format-select').value || 'mp3', // Format for final download
stream: true, stream: true,
speed: speed, speed: speed,
return_download_link: true, return_download_link: true,