Labyricorn/voicebox
1
0
Fork 0
mirror of https://github.com/jamiepine/voicebox.git synced 2026-09-29 07:05:14 -07:00
Code Issues Packages Projects Releases Wiki Activity

Fix runaway MLX Qwen audio chunks (#964)

Browse Source 
* fix runaway MLX Qwen audio chunks

* test: tighten runaway retry coverage

---------

Co-authored-by: huanghua01 <[email protected]>
This commit is contained in:
AirHua
2026-07-26 23:33:09 -07:00
committed by Jamie Pine
co-authored by huanghua01
parent 0d31687638
commit eab3d45192
7 changed files with 262 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files
backend/services/generation.py
+18 -4
View File
@@ -48,8 +48,13 @@ async def run_generation(
This is the single entry point for all background generation work.
It is designed to be enqueued via ``services.task_queue.enqueue_generation``.
"""
from ..backends import engine_needs_trim, get_tts_backend_for_engine, load_engine_model
from ..utils.audio import normalize_audio, save_audio, trim_tts_output
from ..backends import (
engine_needs_trim,
engine_retries_runaway,
get_tts_backend_for_engine,
load_engine_model,
)
from ..utils.audio import has_tts_runaway, normalize_audio, save_audio, trim_tts_output
from ..utils.chunked_tts import generate_chunked
task_manager = get_task_manager()
@@ -72,12 +77,14 @@ async def run_generation(
await history.update_generation_status(generation_id, "generating", bg_db)
trim_fn = trim_tts_output if engine_needs_trim(engine) else None
runaway_detector = has_tts_runaway if engine_retries_runaway(engine) else None
gen_kwargs: dict = dict(
language=language,
seed=seed if mode != "regenerate" else None,
instruct=instruct,
trim_fn=trim_fn,
runaway_detector=runaway_detector,
)
if max_chunk_chars is not None:
gen_kwargs["max_chunk_chars"] = max_chunk_chars
@@ -267,8 +274,13 @@ async def generate_audio_sync(
normalize, then encodes in-memory via :func:`tts.audio_to_wav_bytes`
(same helper ``/generate/stream`` uses).
"""
from ..backends import engine_needs_trim, get_tts_backend_for_engine, load_engine_model
from ..utils.audio import normalize_audio, trim_tts_output
from ..backends import (
engine_needs_trim,
engine_retries_runaway,
get_tts_backend_for_engine,
load_engine_model,
)
from ..utils.audio import has_tts_runaway, normalize_audio, trim_tts_output
from ..utils.chunked_tts import generate_chunked
from . import tts
@@ -287,12 +299,14 @@ async def generate_audio_sync(
bg_db.close()
trim_fn = trim_tts_output if engine_needs_trim(engine) else None
runaway_detector = has_tts_runaway if engine_retries_runaway(engine) else None
gen_kwargs: dict = dict(
language=language,
seed=seed,
instruct=instruct,
trim_fn=trim_fn,
runaway_detector=runaway_detector,
)
if max_chunk_chars is not None:
gen_kwargs["max_chunk_chars"] = max_chunk_chars