Error: API returned: sample_rate_hertz (16000) about googlelanguager HOT 11 CLOSED

Sorry, one remaining issue - I tried a 15 minutes wav file and it keeps coming back with the API Data failed message. I have read some previous closed threads who had a similar issue. What's the best way to overcome this?

async <- gl_speech("gs://6pokemon/buzzfeed.wav", sampleRateHertz = 44100L, asynch = TRUE)
result <- gl_speech_op(async)

Error in data_parse_function(req$content, ...) : object 'out' not found
Warning message:
In call_api() :
  API Data failed to parse.  Returning parsed from JSON content.
                    Use this to test against your data_parse_function.

from googlelanguager.

Everything is working fine now! Thanks Mark!

from googlelanguager.

Try again now with the latest version (and thanks for the test file)

gl_speech("gs://6pokemon/buzzfeed.wav", sampleRateHertz = 44100L, asynch = TRUE)
gl_speech_op(.Last.value)
2018-03-22 16:44:22 -- - started at 2018-03-22T15:39:25.442136Z - last update: 2018-03-22T15:44:09.817086Z
## Send to gl_speech_op() for status
## 3081429970242452248
gl_speech_op(.Last.value)
2018-03-22 16:45:11 -- Asychronous transcription finished.
....

It took about 5 mins to transcribe

from googlelanguager.

It worked magically and yes, I can confirm that it has taken 5 minutes to convert a 15-minutes wav file.

from googlelanguager.

Party :)

from googlelanguager.

Add it via the sampleRateHertz argument e.g.

gl_speech(test_audio, sampleRateHertz = 48000L)$transcript

from googlelanguager.

Thanks Mark.

I found the problem had something to do with the duration of the sound file - it's over 1 minute. So I have uploaded the file to the cloud bucket and put the link into first parameter. I have listed the syntax below and the file is in public. At the moment, the return is "NULL" - very interesting.

gl_speech("gs://6pokemon/gen.wav", encoding = "LINEAR16", sampleRateHertz = 48000L, asynch = TRUE)$transcript
gl_speech("gs://6pokemon/gen.flac", encoding = "FLAC", sampleRateHertz = 48000L, asynch = TRUE)$transcript

Have you got any examples trying to translate a wav/flac file, which is more than 2 or 3 minutes - like a song's length)?

from googlelanguager.

I just fixed that a second ago in this commit e364110 , if you download the latest GitHub version - for long running it should return an async object, which you then feed into gl_speech_op() that will return your data once finished

from googlelanguager.

That is setting it back to what is documented here http://code.markedmondson.me/googleLanguageR/articles/speech.html I think it got overwritten in a recent pull, sorry about that.

from googlelanguager.

Hello, Im trying to run a wav fileand I'm continuously getting this error. What to do?
Error: API returned: Must use single channel (mono) audio, but WAV header indicates 2 channels

from googlelanguager.

@AdhishiP please don't comment on closed issues, open a new one, although in your case it looks like this issue may help you: #34

from googlelanguager.