any support for fine-tune audio data longer than 1 minute?

[Jack-Lin-gif]

What is your question?

For finetuning my model, should I prepare audio data less than 15s? I have lots of audios longer than 1 minute, should I split them manually, or there are other convenient ways? Can I use the vad model during fine-tune process?

What's your environment?

• OS (Linux):
• FunASR Version (1.0.0):
• ModelScope Version (1.11.0):
• PyTorch Version (2.0.0):
• How you installed funasr (pip):
• Python version:
• GPU (4090)
• CUDA/cuDNN version (cuda11.7):
• Docker version (funasr-runtime-sdk-cpu-0.4.1)

[LauraGPT]

We would support it soon via flash-attn