Add wav_test.zip (aishell-1 test + kespeech with references)

Add synthesized_data.zip (JSONL training data)

Remove synthesized_data_v3_jsonl.zip (superseded by synthesized_data.zip)

Add wav.zip (kespeech, aishell-1, aishell-4, waihu audio)

Add finetuned_qwen3asr.zip (full Qwen3-ASR training checkpoint)

Add ver3.zip (SpeechLM error corrector adapter)

Add synthesized_data_v3 JSONL bundle (data/*.jsonl)

Add ver2.zip

Add ver1.zip

initial commit