I got the below error when I ran recognition with the hundreds of audio input stream. Those files chunked by 16 concurrent requests - run in the 16 threads concurrently.
Is it related with the memory size of the GPU or just application error?
LOG ([5.5.0~2-2abe]:TryAddChunkToNextBatchDeepCopy():cuda-online-pipeline-dynamic-batcher.cc:94) All decoding channels are in use. Consider increasing --num-channels
I got the below error when I ran recognition with the hundreds of audio input stream. Those files chunked by 16 concurrent requests - run in the 16 threads concurrently. Is it related with the memory size of the GPU or just application error?
LOG ([5.5.0~2-2abe]:TryAddChunkToNextBatchDeepCopy():cuda-online-pipeline-dynamic-batcher.cc:94) All decoding channels are in use. Consider increasing --num-channels