fix: remove --max-seq-len from voiceclone command, fix Dockerfile comment

--max-seq-len is not in the current :latest image's openai_server.py (upstream
added it after v5 was built). Remove it from voiceclone command so the
container starts; upstream default (4096) applies. Will be re-added explicitly
once v6 image is built with current upstream.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

Dockerfile

@@ -20,7 +20,7 @@ ARG FASTER_QWEN3_TTS_REF=main
 RUN git clone --depth 1 --branch ${FASTER_QWEN3_TTS_REF} \
     https://github.com/andimarafioti/faster-qwen3-tts.git /app
 
-# Apply DGX Spark patches (max-seq-len support for long reference audio)
+# Apply DGX Spark patches (non_streaming_mode=True, per-voice temperature/top_k/top_p)
 COPY patches/openai_server.patch /tmp/
 RUN cd /app && git apply /tmp/openai_server.patch || true
 

docker/docker-compose.yml

@@ -41,7 +41,6 @@ services:
       --model /models/Qwen3-TTS
       --voices /config/voices.json
       --port 8000
-      --max-seq-len 2048
       "
     deploy:
       resources: