From 58e0016925fcfe9e352ba98220b964b86b2d0ab6 Mon Sep 17 00:00:00 2001 From: Your Name Date: Wed, 23 Apr 2025 00:52:04 -0700 Subject: [PATCH] fix prepare.sh --- egs/speech_llm/ASR_LLM/prepare.sh | 13 ++++++++----- egs/speech_llm/ASR_LLM/shared | 1 + 2 files changed, 9 insertions(+), 5 deletions(-) create mode 120000 egs/speech_llm/ASR_LLM/shared diff --git a/egs/speech_llm/ASR_LLM/prepare.sh b/egs/speech_llm/ASR_LLM/prepare.sh index 6f5ed5448..8ca3c1c36 100755 --- a/egs/speech_llm/ASR_LLM/prepare.sh +++ b/egs/speech_llm/ASR_LLM/prepare.sh @@ -7,6 +7,9 @@ set -eou pipefail stage=0 stop_stage=0 + +. shared/parse_options.sh || exit 1 + # All files generated by this script are saved in "data". # You can safely remove "data" and rerun this script to regenerate it. mkdir -p data @@ -23,7 +26,7 @@ if [ $stage -le 0 ] && [ $stop_stage -ge 0 ]; then # pip install huggingface_hub['cli'] # for aishell 1 - huggingface-cli download --local-dir data yuekai/aishell_whisper_fbank_lhotse + huggingface-cli download --repo-type dataset --local-dir data yuekai/aishell_whisper_fbank_lhotse fi @@ -31,9 +34,9 @@ if [ $stage -le 1 ] && [ $stop_stage -ge 1 ]; then log "stage 1: Download whisper-large-v2 multi-hans-zh fbank feature from huggingface" # for multi-hans-zh - huggingface-cli download --local-dir data/fbank yuekai/wenetspeech_whisper_fbank_lhotse - huggingface-cli download --local-dir data/fbank yuekai/multi_hans_zh_whisper_fbank_lhotse - huggingface-cli download --local-dir data/fbank yuekai/alimeeting_aishell4_training_whisper_fbank_lhotse + huggingface-cli download --repo-type dataset --local-dir data/fbank yuekai/wenetspeech_whisper_fbank_lhotse + huggingface-cli download --repo-type dataset --local-dir data/fbank yuekai/multi_hans_zh_whisper_fbank_lhotse + huggingface-cli download --repo-type dataset --local-dir data/fbank yuekai/alimeeting_aishell4_training_whisper_fbank_lhotse fi if [ $stage -le 2 ] && [ $stop_stage -ge 2 ]; then @@ -41,6 +44,6 @@ if [ $stage -le 2 ] && [ $stop_stage -ge 2 ]; then # for speechio test sets mkdir data_speechio - huggingface-cli download --local-dir data_speechio yuekai/icefall_asr_speechio + huggingface-cli download --repo-type model --local-dir data_speechio yuekai/icefall_asr_speechio mv data_speechio/fbank/* data/fbank fi diff --git a/egs/speech_llm/ASR_LLM/shared b/egs/speech_llm/ASR_LLM/shared new file mode 120000 index 000000000..4cbd91a7e --- /dev/null +++ b/egs/speech_llm/ASR_LLM/shared @@ -0,0 +1 @@ +../../../icefall/shared \ No newline at end of file