Deepspeech: OSError: input_filepath tsv_dir/clips/common_voice_en_18885784.mp3 does not exist.

[arajan10]

i am confused as to what is happening!

Loading TSV file: /home/arajan/Documents/Abhishek Rajan/Githubrepo/DeepSpeech-master/tsv_dir/train.tsv
Saving new DeepSpeech-formatted CSV file to: tsv_dir/clips/train.csv
Importing mp3 files...
Traceback (most recent call last):
File "bin/import_cv2.py", line 164, in
_preprocess_data(PARAMS.tsv_dir, AUDIO_DIR, label_filter_fun, PARAMS.space_after_every_character)
File "bin/import_cv2.py", line 43, in _preprocess_data
_maybe_convert_set(input_tsv, audio_dir, label_filter, space_after_every_character)
File "bin/import_cv2.py", line 99, in _maybe_convert_set
for i, _ in enumerate(pool.imap_unordered(one_sample, samples), start=1):
File "/usr/lib/python3.7/multiprocessing/pool.py", line 748, in next
raise value
File "/usr/lib/python3.7/multiprocessing/pool.py", line 121, in worker
result = (True, func(args, *kwds))
File "bin/import_cv2.py", line 70, in one_sample
_maybe_convert_wav(mp3_filename, wav_filename)
File "bin/import_cv2.py", line 133, in _maybe_convert_wav
transformer.build(mp3_filename, wav_filename)
File "/home/arajan/tmp/deepspeech-train-venv/lib/python3.7/site-packages/sox/transform.py", line 412, in build
file_info.validate_input_file(input_filepath)
File "/home/arajan/tmp/deepspeech-train-venv/lib/python3.7/site-packages/sox/file_info.py", line 215, in validate_input_file
"input_filepath {} does not exist.".format(input_filepath)
OSError: input_filepath tsv_dir/clips/common_voice_en_18885784.mp3 does not exist.

[kdavis-mozilla]

Does the file

...input_filepath tsv_dir/clips/common_voice_en_18885784.mp3

exist?

From the error message it looks like it does not.

In a related question, what command line arguments did you use?

[arajan10]

hi kdavis i am using
bin/import_cv2.py tsv_dir
just to train the data .
tsv_dir contains train.csv , test.csv , dev.tsv
tsv_dir/clips contains audio files

[kdavis-mozilla]

What happens if you use an absolute path for 'tsv_dir'

[arajan10]

command used
bin/import_cv2.py -h --/home/arajan/Documents/Abhishek Rajan/Githubrepo/DeepSpeech-master/tsv_dir

output:-

usage: import_cv2.py [-h] [--audio_dir AUDIO_DIR]
[--filter_alphabet FILTER_ALPHABET] [--normalize]
[--space_after_every_character]
tsv_dir

Import CommonVoice v2.0 corpora

positional arguments:
tsv_dir Directory containing tsv files

optional arguments:
-h, --help show this help message and exit
--audio_dir AUDIO_DIR
Directory containing the audio clips - defaults to
"/clips"
--filter_alphabet FILTER_ALPHABET
Exclude samples with characters not in provided
alphabet
--normalize Converts diacritic characters to their base ones
--space_after_every_character
To help transcript join by white space

[kdavis-mozilla]

Why do you include a leading "--" on your path? It's not needed.

[arajan10]

ok ran the same command without -- .
same out put!
am i doing some thing wrong?

[kdavis-mozilla]

And don't include "-h" too.

[arajan10]

ohh ok
bin/import_cv2.py /home/arajan/Documents/Abhishek_Rajan/Githubrepo/DeepSpeech-master/tsv_dir
was the command!

Loading TSV file: /home/arajan/Documents/Abhishek_Rajan/Githubrepo/DeepSpeech-master/tsv_dir/train.tsv
Saving new DeepSpeech-formatted CSV file to: /home/arajan/Documents/Abhishek_Rajan/Githubrepo/DeepSpeech-master/tsv_dir/clips/train.csv
Importing mp3 files...
Traceback (most recent call last):
File "bin/import_cv2.py", line 164, in
_preprocess_data(PARAMS.tsv_dir, AUDIO_DIR, label_filter_fun, PARAMS.space_after_every_character)
File "bin/import_cv2.py", line 43, in _preprocess_data
_maybe_convert_set(input_tsv, audio_dir, label_filter, space_after_every_character)
File "bin/import_cv2.py", line 99, in _maybe_convert_set
for i, _ in enumerate(pool.imap_unordered(one_sample, samples), start=1):
File "/usr/lib/python3.7/multiprocessing/pool.py", line 748, in next
raise value
File "/usr/lib/python3.7/multiprocessing/pool.py", line 121, in worker
result = (True, func(args, *kwds))
File "bin/import_cv2.py", line 70, in one_sample
_maybe_convert_wav(mp3_filename, wav_filename)
File "bin/import_cv2.py", line 133, in _maybe_convert_wav
transformer.build(mp3_filename, wav_filename)
File "/home/arajan/tmp/deepspeech-train-venv/lib/python3.7/site-packages/sox/transform.py", line 412, in build
file_info.validate_input_file(input_filepath)
File "/home/arajan/tmp/deepspeech-train-venv/lib/python3.7/site-packages/sox/file_info.py", line 215, in validate_input_file
"input_filepath {} does not exist.".format(input_filepath)
OSError: input_filepath /home/arajan/Documents/Abhishek_Rajan/Githubrepo/DeepSpeech-master/tsv_dir/clips/common_voice_en_18885788.mp3 does not exist.

[kdavis-mozilla]

Now back to my old question. Does

/home/arajan/Documents/Abhishek_Rajan/Githubrepo/DeepSpeech-master/tsv_dir/clips/common_voice_en_18885788.mp3

exist? And do you have permissions to read it?

[arajan10]

if i gave system common_voice_en_18885788.mp3 .it asks for common_voice_en_18885789.mp3 .
i went on and now
OSError: input_filepath /home/arajan/Documents/Abhishek_Rajan/Githubrepo/DeepSpeech-master/tsv_dir/clips/common_voice_en_18885791.mp3 does not exist.

[kdavis-mozilla]

No. Does the file

/home/arajan/Documents/Abhishek_Rajan/Githubrepo/DeepSpeech-master/tsv_dir/clips/common_voice_en_18885788.mp3

exist? And do you have permissions to read it?

[arajan10]

yes

[kdavis-mozilla]

Is that yes to both questions?

[arajan10]

yes that resolved the issue!

[lock[bot]]

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.