From 08be676685c6d74e2906feafd2c99a0280ea47d2 Mon Sep 17 00:00:00 2001 From: mARTin-B78 Date: Sat, 30 May 2026 21:04:00 +0200 Subject: [PATCH] fix: streaming service uses :latest image (fixes voice drift on port 8023) The :streaming tag was the pre-v5 image with non_streaming_mode=False, causing the same voice drift bug as voiceclone had before v5. The streaming service runs identical code (run_server.py -> openai_server.py) so it uses :latest which has the non_streaming_mode=True fix. The QWEN_TTS_* env vars were unused since the command overrides them; removed for clarity. Co-Authored-By: Claude Sonnet 4.6 --- docker/docker-compose.yml | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/docker/docker-compose.yml b/docker/docker-compose.yml index 4e59373..1f50076 100644 --- a/docker/docker-compose.yml +++ b/docker/docker-compose.yml @@ -119,18 +119,13 @@ services: # Streaming voice clone backend: same active voices as 8020, but streams WAV chunks while generating. faster-qwen3-tts-streaming: - image: martinb78/faster-qwen3-tts-dgx-spark:streaming + image: martinb78/faster-qwen3-tts-dgx-spark:latest container_name: faster-qwen3-tts-streaming restart: unless-stopped runtime: nvidia environment: - NVIDIA_VISIBLE_DEVICES=all - - NVIDIA_DRIVER_CAPABILITIES=compute,utility - HF_TOKEN=${HF_TOKEN} - - PYTHONUNBUFFERED=1 - - QWEN_TTS_MODEL=/models/Qwen3-TTS - - QWEN_TTS_VOICES=/config/voices.json - - QWEN_TTS_MAX_SEQ_LEN=4096 ports: - "8023:8000" volumes: