[Feat] Add benchmarks for Qwen3-TTS Base/VoiceDesign Model (#2411)

Signed-off-by: Jiahui Sun <jhsun2020@gmail.com>

benchmarks/qwen3-tts/README.md

@@ -32,6 +32,9 @@ bash run_benchmark.sh --hf-only
 # Use a different model (e.g. 1.7B)
 MODEL=Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice bash run_benchmark.sh --async-only
 
+# Use a Voice Clone model
+MODEL=Qwen/Qwen3-TTS-12Hz-1.7B-Base TASK_TYPE=Base bash run_benchmark.sh --async-only
+
 # Use bs16 config for higher throughput
 STAGE_CONFIG=vllm_omni/configs/qwen3_tts_bs16.yaml bash run_benchmark.sh --async-only
 

benchmarks/qwen3-tts/run_benchmark.sh

@@ -23,6 +23,9 @@
 #   # Use 1.7B model:
 #   MODEL=Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice bash run_benchmark.sh --async-only
 #
+#   # Use Voice Clone model
+#   MODEL=Qwen/Qwen3-TTS-12Hz-1.7B-Base TASK_TYPE=Base bash run_benchmark.sh --async-only
+#
 #   # Use batch_size=4 config:
 #   STAGE_CONFIG=vllm_omni/configs/qwen3_tts_bs4.yaml bash run_benchmark.sh --async-only
 #
@@ -35,6 +38,7 @@
 #   GPU_MEM_TALKER   - gpu_memory_utilization for talker stage (default: 0.3)
 #   GPU_MEM_CODE2WAV - gpu_memory_utilization for code2wav stage (default: 0.2)
 #   STAGE_CONFIG     - Path to stage config YAML (default: configs/qwen3_tts_bs1.yaml)
+#   TASK_TYPE        - Task type: CustomVoice, VoiceDesign, Base (default: CustomVoice)
 
 set -euo pipefail
 
@@ -53,6 +57,7 @@ NUM_WARMUPS="${NUM_WARMUPS:-3}"
 STAGE_CONFIG="${STAGE_CONFIG:-vllm_omni/configs/qwen3_tts_bs1.yaml}"
 RESULT_DIR="${SCRIPT_DIR}/results"
 TIMESTAMP="$(date +%Y%m%d_%H%M%S)"
+TASK_TYPE="${TASK_TYPE:-CustomVoice}"
 
 # Parse args
 RUN_ASYNC=true
@@ -77,6 +82,7 @@ echo " Concurrency:  ${CONCURRENCY}"
 echo " Port:         ${PORT}"
 echo " Stage config: ${STAGE_CONFIG}"
 echo " Results:      ${RESULT_DIR}"
+echo " Task type:    ${TASK_TYPE}"
 echo "============================================================"
 
 # Prepare stage config with correct GPU device and memory settings
@@ -195,7 +201,8 @@ run_bench() {
         --max-concurrency ${conc_args} \
         --num-warmups "${NUM_WARMUPS}" \
         --config-name "${config_name}" \
-        --result-dir "${RESULT_DIR}"
+        --result-dir "${RESULT_DIR}" \
+        --task-type "${TASK_TYPE}"
 
     stop_server
 
@@ -222,7 +229,8 @@ if [ "${RUN_HF}" = true ]; then
         --num-warmups "${NUM_WARMUPS}" \
         --gpu-device "${GPU_DEVICE}" \
         --config-name "hf_transformers" \
-        --result-dir "${RESULT_DIR}"
+        --result-dir "${RESULT_DIR}" \
+        --task-type "${TASK_TYPE}"
 
     # Allow GPU memory to settle
     sleep 5

benchmarks/qwen3-tts/transformers/bench_tts_hf.py

@@ -38,6 +38,10 @@ PROMPTS = [
     "It was a dark and stormy night when the old lighthouse keeper heard a knock at the door.",
 ]
 
+REF_AUDIO = "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-TTS-Repo/clone_2.wav"
+REF_TEXT = "Okay. Yeah. I resent you. I love you. I respect you. But you know what? You blew it! And thanks to you."
+INSTRUCT = "Speak in an incredulous tone, but with a hint of panic beginning to creep into your voice."
+
 
 @dataclass
 class BenchmarkResult:
@@ -75,6 +79,29 @@ class BenchmarkResult:
     per_request: list = field(default_factory=list)
 
 
+def generate_audio(model, prompt: str, args):
+    if args.task_type == "Base":
+        return model.generate_voice_clone(
+            text=prompt,
+            language=args.language,
+            ref_audio=REF_AUDIO,
+            ref_text=REF_TEXT,
+        )
+
+    if args.task_type == "VoiceDesign":
+        return model.generate_voice_design(
+            text=prompt,
+            language=args.language,
+            instruct=INSTRUCT,
+        )
+
+    return model.generate_custom_voice(
+        text=prompt,
+        language=args.language,
+        speaker=args.voice,
+    )
+
+
 def run_benchmark(args):
     from qwen_tts import Qwen3TTSModel
 
@@ -95,11 +122,7 @@ def run_benchmark(args):
         print(f"Warming up with {args.num_warmups} requests...")
         for i in range(args.num_warmups):
             p = PROMPTS[i % len(PROMPTS)]
-            wavs, sr = model.generate_custom_voice(
-                text=p,
-                language=args.language,
-                speaker=args.voice,
-            )
+            wavs, sr = generate_audio(model, p, args)
         # Sync GPU
         torch.cuda.synchronize(device)
         print("Warmup done.")
@@ -124,11 +147,7 @@ def run_benchmark(args):
             torch.cuda.synchronize(device)
             st = time.perf_counter()
 
-            wavs, sr = model.generate_custom_voice(
-                text=prompt,
-                language=args.language,
-                speaker=args.voice,
-            )
+            wavs, sr = generate_audio(model, prompt, args)
 
             torch.cuda.synchronize(device)
             elapsed = time.perf_counter() - st
@@ -268,6 +287,7 @@ def parse_args():
     parser.add_argument("--gpu-device", type=int, default=0)
     parser.add_argument("--voice", type=str, default="Vivian")
     parser.add_argument("--language", type=str, default="English")
+    parser.add_argument("--task-type", type=str, default="CustomVoice", choices=["CustomVoice", "VoiceDesign", "Base"])
     parser.add_argument(
         "--config-name", type=str, default="hf_transformers", help="Label for this config (used in filenames)"
     )

benchmarks/qwen3-tts/vllm_omni/bench_tts_serve.py

@@ -37,6 +37,9 @@ PROMPTS = [
     "Could you please turn down the music a little bit, I'm trying to concentrate on my work.",
     "It was a dark and stormy night when the old lighthouse keeper heard a knock at the door.",
 ]
+REF_AUDIO = "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-TTS-Repo/clone_2.wav"
+REF_TEXT = "Okay. Yeah. I resent you. I love you. I respect you. But you know what? You blew it! And thanks to you."
+INSTRUCT = "Speak in an incredulous tone, but with a hint of panic beginning to creep into your voice."
 
 
 @dataclass
@@ -93,22 +96,39 @@ def pcm_bytes_to_duration(num_bytes: int, sample_rate: int = 24000, sample_width
     return num_samples / sample_rate
 
 
+def create_payload(
+    prompt: str, task_type: str = "CustomVoice", voice: str = "vivian", language: str = "English"
+) -> dict:
+    payload = {
+        "input": prompt,
+        "language": language,
+        "stream": True,
+        "response_format": "pcm",
+        "task_type": task_type,
+    }
+
+    if task_type == "Base":
+        payload["ref_audio"] = REF_AUDIO
+        payload["ref_text"] = REF_TEXT
+    elif task_type == "CustomVoice":
+        payload["voice"] = voice
+    elif task_type == "VoiceDesign":
+        payload["instructions"] = INSTRUCT
+
+    return payload
+
+
 async def send_tts_request(
     session: aiohttp.ClientSession,
     api_url: str,
     prompt: str,
+    task_type: str = "CustomVoice",
     voice: str = "vivian",
     language: str = "English",
     pbar: tqdm | None = None,
 ) -> RequestResult:
     """Send a streaming TTS request and measure latency metrics."""
-    payload = {
-        "input": prompt,
-        "voice": voice,
-        "language": language,
-        "stream": True,
-        "response_format": "pcm",
-    }
+    payload = create_payload(prompt, task_type, voice, language)
 
     result = RequestResult(prompt=prompt)
     st = time.perf_counter()
@@ -153,6 +173,7 @@ async def run_benchmark(
     num_prompts: int,
     max_concurrency: int,
     num_warmups: int = 3,
+    task_type: str = "CustomVoice",
     voice: str = "vivian",
     language: str = "English",
 ) -> BenchmarkResult:
@@ -175,7 +196,7 @@ async def run_benchmark(
         warmup_tasks = []
         for i in range(num_warmups):
             prompt = PROMPTS[i % len(PROMPTS)]
-            warmup_tasks.append(send_tts_request(session, api_url, prompt, voice, language))
+            warmup_tasks.append(send_tts_request(session, api_url, prompt, task_type, voice, language))
         await asyncio.gather(*warmup_tasks)
         print("  Warmup done.")
 
@@ -189,7 +210,7 @@ async def run_benchmark(
 
     async def limited_request(prompt):
         async with semaphore:
-            return await send_tts_request(session, api_url, prompt, voice, language, pbar)
+            return await send_tts_request(session, api_url, prompt, task_type, voice, language, pbar)
 
     start_time = time.perf_counter()
     tasks = [asyncio.create_task(limited_request(p)) for p in request_prompts]
@@ -306,6 +327,7 @@ async def main(args):
             num_prompts=args.num_prompts,
             max_concurrency=concurrency,
             num_warmups=args.num_warmups,
+            task_type=args.task_type,
             voice=args.voice,
             language=args.language,
         )
@@ -334,6 +356,7 @@ def parse_args():
         "--max-concurrency", type=int, nargs="+", default=[1, 4, 10], help="Concurrency levels to test"
     )
     parser.add_argument("--num-warmups", type=int, default=3)
+    parser.add_argument("--task-type", type=str, default="CustomVoice", choices=["CustomVoice", "VoiceDesign", "Base"])
     parser.add_argument("--voice", type=str, default="vivian")
     parser.add_argument("--language", type=str, default="English")
     parser.add_argument(