pytorch
torchaudio.load not loading all the frames
🐛 Describe the bug
torchaudio.load not loading all the frames in the latest version(2.2.1).
Example audio can be downloaded from here
import torchaudio
file = "harddisk_operation.wav"
audio, sr = torchaudio.load(file)
print("num_frames:", audio.shape[-1])
For the above code,
torchaudio 2.2.1 gives an output - num_frames: 832779
torchaudio 2.0.0 gives an output - num_frames: 1110372
Versions
Collecting environment information... PyTorch version: 2.2.1+cu121 Is debug build: False CUDA used to build PyTorch: 12.1 ROCM used to build PyTorch: N/A
OS: Ubuntu 20.04.2 LTS (x86_64) GCC version: (Ubuntu 9.4.0-1ubuntu1~20.04.2) 9.4.0 Clang version: Could not collect CMake version: Could not collect Libc version: glibc-2.31
Python version: 3.10.6 (main, Oct 24 2022, 16:07:47) [GCC 11.2.0] (64-bit runtime) Python platform: Linux-5.15.0-94-generic-x86_64-with-glibc2.31 Is CUDA available: True CUDA runtime version: Could not collect CUDA_MODULE_LOADING set to: LAZY GPU models and configuration: GPU 0: NVIDIA RTX A6000 GPU 1: NVIDIA RTX A6000 GPU 2: NVIDIA RTX A6000 GPU 3: NVIDIA RTX A6000 GPU 4: NVIDIA RTX A6000 GPU 5: NVIDIA RTX A6000 GPU 6: NVIDIA RTX A6000 GPU 7: NVIDIA RTX A6000
Nvidia driver version: 525.147.05 cuDNN version: Could not collect HIP runtime version: N/A MIOpen runtime version: N/A Is XNNPACK available: True
CPU: Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Byte Order: Little Endian Address sizes: 43 bits physical, 48 bits virtual CPU(s): 256 On-line CPU(s) list: 0-255 Thread(s) per core: 2 Core(s) per socket: 64 Socket(s): 2 NUMA node(s): 2 Vendor ID: AuthenticAMD CPU family: 23 Model: 49 Model name: AMD EPYC 7662 64-Core Processor Stepping: 0 Frequency boost: enabled CPU MHz: 2630.056 CPU max MHz: 2154.2959 CPU min MHz: 1500.0000 BogoMIPS: 3999.82 Virtualization: AMD-V L1d cache: 4 MiB L1i cache: 4 MiB L2 cache: 64 MiB L3 cache: 512 MiB NUMA node0 CPU(s): 0-63,128-191 NUMA node1 CPU(s): 64-127,192-255 Vulnerability Gather data sampling: Not affected Vulnerability Itlb multihit: Not affected Vulnerability L1tf: Not affected Vulnerability Mds: Not affected Vulnerability Meltdown: Not affected Vulnerability Mmio stale data: Not affected Vulnerability Retbleed: Mitigation; untrained return thunk; SMT enabled with STIBP protection Vulnerability Spec rstack overflow: Mitigation; safe RET Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization Vulnerability Spectre v2: Mitigation; Retpolines, IBPB conditional, STIBP always-on, RSB filling, PBRSB-eIBRS Not affected Vulnerability Srbds: Not affected Vulnerability Tsx async abort: Not affected Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc cpuid extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb cat_l3 cdp_l3 hw_pstate ssbd mba ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr rdpru wbnoinvd amd_ppin arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif v_spec_ctrl umip rdpid overflow_recov succor smca sme sev sev_es
Versions of relevant libraries: [pip3] numpy==1.26.4 [pip3] torch==2.2.1 [pip3] torchaudio==2.2.1 [pip3] triton==2.2.0 [conda] numpy 1.26.4 pypi_0 pypi [conda] torch 2.2.1 pypi_0 pypi [conda] torchaudio 2.2.1 pypi_0 pypi [conda] triton 2.2.0 pypi_0 pypi
I observed a similar problem with a .flac file that gave 661794 samples in 2.0.0 (which matches what soxi and ffprobe report for that file), but only 656640 samples in 2.3.1. In 2.4.0, the length is correct again. Can you also check in 2.4.0?
/edit: My 2.4.0 environment has ffmpeg 4.3, while the other has ffmpeg 4.2.2. This could also make a difference, although for both ffmpeg versions, ffprobe -i gtzan_rock_00014.flac -show_entries format=duration -v quiet -of csv="p=0" shows the correct length of 30.013333, which matches 661794 samples at 22050 Hz.
/editedit: I can confirm it is related to the installed version of ffmpeg. When using conda, with pkgs/main::ffmpeg-4.2.2-h20bf706_0, I got the wrong length, and with pytorch::ffmpeg-4.3-hf484d3e_0 I got the correct one. Tested with python -c "import torchaudio; print(torchaudio.load('gtzan_rock_00014.flac')[0].shape)".
