Initial commit: FunASR Speech Recognition Toolkit

Add complete FunASR codebase including models, runtime, and documentation.

examples/industrial_data_pretraining/qwen_audio/demo.py

@@ -0,0 +1,16 @@
+#!/usr/bin/env python3
+# -*- encoding: utf-8 -*-
+# Copyright FunASR (https://github.com/alibaba-damo-academy/FunASR). All Rights Reserved.
+#  MIT License  (https://opensource.org/licenses/MIT)
+
+# To install requirements: pip3 install -U "funasr[llm]"
+
+from funasr import AutoModel
+
+model = AutoModel(model="Qwen-Audio")
+
+audio_in = "https://github.com/QwenLM/Qwen-Audio/raw/main/assets/audio/1272-128104-0000.flac"
+prompt = "<|startoftranscription|><|en|><|transcribe|><|en|><|notimestamps|><|wo_itn|>"
+
+res = model.generate(input=audio_in, prompt=prompt)
+print(res)

examples/industrial_data_pretraining/qwen_audio/demo_chat.py

@@ -0,0 +1,24 @@
+#!/usr/bin/env python3
+# -*- encoding: utf-8 -*-
+# Copyright FunASR (https://github.com/alibaba-damo-academy/FunASR). All Rights Reserved.
+#  MIT License  (https://opensource.org/licenses/MIT)
+
+# To install requirements: pip3 install -U "funasr[llm]"
+
+from funasr import AutoModel
+
+model = AutoModel(model="Qwen/Qwen-Audio-Chat")
+
+audio_in = "https://github.com/QwenLM/Qwen-Audio/raw/main/assets/audio/1272-128104-0000.flac"
+
+# 1st dialogue turn
+prompt = "what does the person say?"
+cache = {"history": None}
+res = model.generate(input=audio_in, prompt=prompt, cache=cache)
+print(res)
+
+
+# 2nd dialogue turn
+prompt = 'Find the start time and end time of the word "middle classes"'
+res = model.generate(input=None, prompt=prompt, cache=cache)
+print(res)

examples/industrial_data_pretraining/qwen_audio/demo_chat_from_local.py

@@ -0,0 +1,29 @@
+#!/usr/bin/env python3
+# -*- encoding: utf-8 -*-
+# Copyright FunASR (https://github.com/alibaba-damo-academy/FunASR). All Rights Reserved.
+#  MIT License  (https://opensource.org/licenses/MIT)
+
+# To install requirements: pip3 install -U "funasr[llm]"
+
+from funasr import AutoModel
+
+model = AutoModel(
+    model="Qwen-Audio-Chat",
+    model_path="/nfs/zhifu.gzf/init_model/qwen/Qwen-Audio-Chat",
+)
+
+audio_in = (
+    "https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr_example_zh.wav"
+)
+
+# 1st dialogue turn
+prompt = "what does the person say?"
+cache = {"history": None}
+res = model.generate(input=audio_in, prompt=prompt, cache=cache)
+print(res)
+
+
+# 2nd dialogue turn
+prompt = 'Find the start time and end time of the word "middle classes"'
+res = model.generate(input=None, prompt=prompt, cache=cache)
+print(res)

examples/industrial_data_pretraining/qwen_audio/demo_from_local.py

@@ -0,0 +1,19 @@
+#!/usr/bin/env python3
+# -*- encoding: utf-8 -*-
+# Copyright FunASR (https://github.com/alibaba-damo-academy/FunASR). All Rights Reserved.
+#  MIT License  (https://opensource.org/licenses/MIT)
+
+# To install requirements: pip3 install -U "funasr[llm]"
+
+from funasr import AutoModel
+
+model = AutoModel(
+    model="Qwen-Audio",
+    model_path="/nfs/zhifu.gzf/init_model/qwen/Qwen-Audio",
+)
+
+audio_in = "https://github.com/QwenLM/Qwen-Audio/raw/main/assets/audio/1272-128104-0000.flac"
+prompt = "<|startoftranscription|><|en|><|transcribe|><|en|><|notimestamps|><|wo_itn|>"
+
+res = model.generate(input=audio_in, prompt=prompt)
+print(res)