add model

.gitattributes

@@ -1,4 +1,5 @@
 *.7z filter=lfs diff=lfs merge=lfs -text
+Qwen3-ASR-1.7B/**/* filter=lfs diff=lfs merge=lfs -text
 *.pdf filter=lfs diff=lfs merge=lfs -text
 *.arrow filter=lfs diff=lfs merge=lfs -text
 *.avro filter=lfs diff=lfs merge=lfs -text

Qwen3-ASR-1.7B/.gitattributes

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11ad7efa24975ee4b0c3c3a38ed18737f0658a5f75a0a96787b576a78a023361
+size 1519

Qwen3-ASR-1.7B/README.md

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5058416891bc47a2051557765997e8c42f8eb78a0e33c3e775bd17d4b0ba4d50
+size 57456

Qwen3-ASR-1.7B/chat_template.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75a8cfca24f00de72d796fbfed6858fc9614ef3dabd8696684cc3bc03a9c58ff
+size 1161

Qwen3-ASR-1.7B/config.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e74a751548b8ad7d7526d29365ad8144c345d8b412b1152d25dc6698452712f
+size 6194

Qwen3-ASR-1.7B/generation_config.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1da527824d81e07118facff437e03f2e24a23311e3bdeb2368973fe77e5f275c
+size 142

Qwen3-ASR-1.7B/merges.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5
+size 1671853

Qwen3-ASR-1.7B/model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4cd1f1a04d90b757dc7f7dd26254e69a013b19e80efe590a83c6a3bde8608d6
+size 4220320824

Qwen3-ASR-1.7B/model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e0b9d9e09e2e0238e7ef3cc8a484ab387e91b90f1900bedf88bc92d7929ccfc
+size 478200688

Qwen3-ASR-1.7B/model.safetensors.index.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f994739fe38e5210b9e3e8ce6c6307315e2ceac3cb630e7b7414d69dce520f60
+size 64821

Qwen3-ASR-1.7B/preprocessor_config.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45e120a4eda2c20c5d7f2ea9354e63536bf35e27aa573fb7cdf78017b378770d
+size 330

Qwen3-ASR-1.7B/tokenizer_config.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4942d005604266809309cabc9f4e9cb89ce855d59b14681fdc0e1cc62ea26c4c
+size 12487

Qwen3-ASR-1.7B/vocab.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910
+size 2776833

example_qwen3_asr_vllm_streaming.py

@@ -34,7 +34,8 @@ import soundfile as sf
 from qwen_asr import Qwen3ASRModel
 
 
-ASR_MODEL_PATH = "Qwen/Qwen3-ASR-1.7B"
+# ASR_MODEL_PATH = "Qwen/Qwen3-ASR-1.7B"
+ASR_MODEL_PATH = "./Qwen3-ASR-1.7B"
 URL_EN = "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_en.wav"
 
 
@@ -92,8 +93,11 @@ def main() -> None:
     # Streaming is vLLM-only and no forced aligner supported.
     asr = Qwen3ASRModel.LLM(
         model=ASR_MODEL_PATH,
-        gpu_memory_utilization=0.8,
-        max_new_tokens=32, # set a small value for streaming
+        gpu_memory_utilization=0.6,
+        max_model_len=4096,
+        max_num_seqs=1,
+        enforce_eager=True,
+        max_new_tokens=32,  # set a small value for streaming
     )
 
     audio_bytes = _download_audio_bytes(URL_EN)

readme.txt

@@ -1,4 +1,2 @@
 conda create -n qwen3-asr python=3.12 -y
 conda activate qwen3-asr
-
-torch==2.12.0