Please provide the following information when requesting support.
Hardware - GPU (A100/A30/T4/V100): T4
Hardware - CPU: x86_64
Operating System: Ubuntu
Riva Version: 2.14.0 (tried 2.8.1too)
TLT Version (if relevant)
How to reproduce the issue ? (This is for errors. Please share the command and the detailed log here)
Using both nvcr.io/nvidia/riva/riva-speech:2.14.0-servicemaker
and nvcr.io/nvidia/riva/riva-speech:2.8.1-servicemaker
Launching
$ docker run --gpus all -it --rm \
-v /home/ubuntu/riva_artifacts:/servicemaker-dev \
-v /home/ubuntu/riva_repo:/data \
--entrypoint="/bin/bash" \
nvcr.io/nvidia/riva/riva-speech:2.14.0-servicemaker
Once inside
# riva-build speech_recognition \
/servicemaker-dev/file.rmir:akey1 /servicemaker-dev/anotherfile.riva:akey1 \
--name=a_name --decoding_vocab=/servicemaker-dev/custom_vocabulary.txt \
--decoder_type=flashlight
Output: [ERROR] /servicemaker-dev/file.rmir exists. Use --force/-f to overwrite
Weird because I guess is supposed to exist, but ok let’s use -f
# riva-build speech_recognition -f \
/servicemaker-dev/file.rmir:akey1 /servicemaker-dev/anotherfile.riva:akey1 \
--name=a_name --decoding_vocab=/servicemaker-dev/custom_vocabulary.txt \
--decoder_type=flashlight
Output:
Traceback (most recent call last):
File "/usr/local/bin/riva-build", line 8, in <module>
sys.exit(build())
File "/usr/local/lib/python3.8/dist-packages/servicemaker/cli/build.py", line 96, in build
nm = model(input_filename, pipeline_config=pipeline_config, encryption_key=input_key)
File "/usr/local/lib/python3.8/dist-packages/servicemaker/readers/tlt.py", line 76, in __init__
read_eff_manifest(self, self.source_file)
File "/usr/local/lib/python3.8/dist-packages/servicemaker/readers/tlt.py", line 65, in read_eff_manifest
manifest = Archive.restore_manifest(restore_path=source_file)
File "<frozen eff.core.archive>", line 646, in restore_manifest
FileNotFoundError: The indicated '/servicemaker-dev/anotherfile.riva' file does not exist