Tests failing

[jpschewe]

I'm new to ruby, so I may be missing something basic. I installed curb and json with "sudo gem install curb; sudo gem install json". Then added "require 'rubygems'" to lib/speech.rb because the new gems weren't found. I then got this:

jpschewe@jon:~/src/speech2text
[master] >rake
(in /home/jpschewe/src/speech2text)
Loaded suite /usr/lib64/ruby/gems/1.8/gems/rake-0.8.7/lib/rake/rake_test_loader
Started
.EEChunk count: 1
convert: samples/chunk-i-like-pickles-0.chunk5.wav to flac

flac 1.2.1, Copyright (C) 2000,2001,2002,2003,2004,2005,2006,2007  Josh Coalson
flac comes with ABSOLUTELY NO WARRANTY.  This is free software, and you are
welcome to redistribute it under certain conditions.  Type `flac' for details.

chunk-i-like-pickles-0.chunk5.wav: wrote 16781 bytes, ratio=0.455
success?
sending chunk of size 0.57...
32891.00/32891.00
{"hypotheses"=>[["eagles eagles eagles", 0.6748087], ["pickles pickles pickles",

 nil], ["eagles pickles pickles", nil], ["eagles eagles pickles", nil], ["eagles eagles giggles", nil], ["pickles pickles eagles", nil], ["michaels pickles pickles", nil], ["giggles pickles pickles", nil], ["eagle's pickles pickles", nil]], "id"=>"b9bf279c980bd505f67120d274fa7f30-1", "status"=>0}
{"hypotheses"=>[["eagles eagles eagles", 0.6748087], ["pickles pickles pickles", nil], ["eagles pickles pickles", nil], ["eagles eagles pickles", nil], ["eagles eagles giggles", nil], ["pickles pickles eagles", nil], ["michaels pickles pickles", nil], ["giggles pickles pickles", nil], ["eagle's pickles pickles", nil]], "id"=>"b9bf279c980bd505f67120d274fa7f30-1", "status"=>0}
F
Finished in 5.471139 seconds.

  1) Error:
test_audio_splitter(SpeechAudioSplitterTest):
NoMethodError: undefined method `first' for nil:NilClass
    /home/jpschewe/src/speech2text/lib/speech/audio_inspector.rb:50:in `initialize'
    /home/jpschewe/src/speech2text/lib/speech/audio_splitter.rb:77:in `new'
    /home/jpschewe/src/speech2text/lib/speech/audio_splitter.rb:77:in `initialize'
    ./test/audio_splitter_test.rb:9:in `new'
    ./test/audio_splitter_test.rb:9:in `test_audio_splitter'

 2) Error:
test_audio_to_text(SpeechAudioToTextTest):
NoMethodError: undefined method `first' for nil:NilClass
    /home/jpschewe/src/speech2text/lib/speech/audio_inspector.rb:50:in `initialize'
    /home/jpschewe/src/speech2text/lib/speech/audio_splitter.rb:77:in `new'
    /home/jpschewe/src/speech2text/lib/speech/audio_splitter.rb:77:in `initialize'
    /home/jpschewe/src/speech2text/lib/speech/audio_to_text.rb:15:in `new'
    /home/jpschewe/src/speech2text/lib/speech/audio_to_text.rb:15:in `to_text'
    ./test/audio_to_text_test.rb:9:in `test_audio_to_text'

  3) Failure:
test_short_audio_clip(SpeechAudioToTextTest) [./test/audio_to_text_test.rb:36]:
<"eagles"> expected but was
<"eagles eagles eagles">.

4 tests, 7 assertions, 1 failures, 2 errors
rake aborted!
Command failed with status (1): [/usr/bin/ruby -I"lib" "/usr/lib64/ruby/gem...]

(See full trace by running task with --trace)

[taf2]

Hi Thanks for checking this out, it's all very much a work in progress. I just pushed a few local edits I have - make sure you have run bundle install.

[jpschewe]

I'm just running out of the source directory. Right know it's unhappy about my input file. Something about 0 bits per sample. I think this is because it's a mono recording. I'm trying to convert an interview for my wife. It's a 2 hour audio file. I've already got it in flac, but didn't see how to easily use that as input. So doing two conversions. Suggestions?

Jon

On Apr 10, 2011, at 0:22, taf2 reply@reply.github.com wrote:

Hi Thanks for checking this out, it's all very much a work in progress. I just pushed a few local edits I have - make sure you have run bundle install.

Reply to this email directly or view it on GitHub:
#1 (comment)

[taf2]

Try this project https://github.com/taf2/audiosplit. It uses the
speech2text code but is a wrapper that splits wave files by word
breaks so it can send smaller chunks of audio to google.... All of
this is a work in progress

On Sunday, April 10, 2011, jpschewe
reply@reply.github.com
wrote:

I'm just running out of the source directory. Right know it's unhappy about my input file. Something about 0 bits per sample. I think this is because it's a mono recording. I'm trying to convert an interview for my wife. It's a 2 hour audio file. I've already got it in flac, but didn't see how to easily use that as input. So doing two conversions. Suggestions?

Jon

On Apr 10, 2011, at 0:22, taf2 reply@reply.github.com wrote:

Hi Thanks for checking this out, it's all very much a work in progress.   I just pushed a few local edits I have - make sure you have run bundle install.

Reply to this email directly or view it on GitHub:
#1 (comment)

Reply to this email directly or view it on GitHub:
#1 (comment)

[jpschewe]

Thanks, I'll try that next. I got the wav file into audacity and only the left track has data. The right track is there, but empty. My guess is that ffmpeg doesn't like this. I'll try mono.

Jon

On Apr 10, 2011, at 9:21, taf2 reply@reply.github.com wrote:

Try this project https://github.com/taf2/audiosplit. It uses the
speech2text code but is a wrapper that splits wave files by word
breaks so it can send smaller chunks of audio to google.... All of
this is a work in progress

On Sunday, April 10, 2011, jpschewe
reply@reply.github.com
wrote:

I'm just running out of the source directory. Right know it's unhappy about my input file. Something about 0 bits per sample. I think this is because it's a mono recording. I'm trying to convert an interview for my wife. It's a 2 hour audio file. I've already got it in flac, but didn't see how to easily use that as input. So doing two conversions. Suggestions?

Jon

On Apr 10, 2011, at 0:22, taf2 reply@reply.github.com wrote:

Hi Thanks for checking this out, it's all very much a work in progress. I just pushed a few local edits I have - make sure you have run bundle install.

Reply to this email directly or view it on GitHub:
#1 (comment)

Reply to this email directly or view it on GitHub:
#1 (comment)

Reply to this email directly or view it on GitHub:
#1 (comment)

[jpschewe]

That didn't work so good. It couldn't find any space in the file,
probably too much background noise, so it appears to have gotten stuck
in a loop.
unk0.chunk0.chunk66.chunk30
chunk
/mnt/external/jpschewe/jen-interview-mono.wav.chunk0.chunk0.chunk0.chunk0.
chunk66.chunk30:0.00 is smaller than min duration: 0.20
create new chunk:
/mnt/external/jpschewe/jen-interview-mono.wav.chunk0.chunk0.chunk0.chunk0.chunk66.chunk31
chunk
/mnt/external/jpschewe/jen-interview-mono.wav.chunk0.chunk0.chunk0.chunk0.chunk66.chunk31:0.11
is smaller than min duration: 0.20
create new chunk:
/mnt/external/jpschewe/jen-interview-mono.wav.chunk0.chunk0.chunk0.chunk0.chunk66.chunk32
chunk
/mnt/external/jpschewe/jen-interview-mono.wav.chunk0.chunk0.chunk0.chunk0.chunk66.chunk32:0.11
is smaller than min duration: 0.20
create new chunk:
/mnt/external/jpschewe/jen-interview-mono.wav.chunk0.chunk0.chunk0.chunk0.chunk66.chunk33

On 04/10/2011 09:21 AM, taf2 wrote:

Try this project https://github.com/taf2/audiosplit. It uses the
speech2text code but is a wrapper that splits wave files by word
breaks so it can send smaller chunks of audio to google.... All of
this is a work in progress

On Sunday, April 10, 2011, jpschewe
reply@reply.github.com
wrote:

I'm just running out of the source directory. Right know it's unhappy about my input file. Something about 0 bits per sample. I think this is because it's a mono recording. I'm trying to convert an interview for my wife. It's a 2 hour audio file. I've already got it in flac, but didn't see how to easily use that as input. So doing two conversions. Suggestions?

Jon

On Apr 10, 2011, at 0:22, taf2 reply@reply.github.com wrote:

Hi Thanks for checking this out, it's all very much a work in progress. I just pushed a few local edits I have - make sure you have run bundle install.

Reply to this email directly or view it on GitHub:

#1 (comment)

Reply to this email directly or view it on GitHub:
#1 (comment)