speech-to-text.rb

def speech_sync_recognize(audio_file_path: nil, language_code: "en-US", profanity_filter: false)
  # [START speech_transcribe_sync]
  # audio_file_path = "Path to file on which to perform speech recognition"

  require "google/cloud/speech"

  speech = Google::Cloud::Speech.speech

  # [START speech_ruby_migration_sync_response]
  audio_file = File.binread audio_file_path
  config     = { encoding:          :LINEAR16,
                 sample_rate_hertz: 16_000,
                 profanity_filter: profanity_filter,
                 language_code:     language_code }
  audio      = { content: audio_file }

  response = speech.recognize config: config, audio: audio

  results = response.results

  alternatives = results.first.alternatives
  alternatives.each do |alternative|
    puts "Transcription: #{alternative.transcript}"
  end
  # [END speech_ruby_migration_sync_response]
  # [END speech_transcribe_sync]
end

audio_file_path = "test-en-US.wav"
speech_sync_recognize(audio_file_path: audio_file_path, language_code: "en-US", profanity_filter: true)

# ffmpeg -i ~/Downloads/IMG_4467.MOV -f s16le -acodec pcm_s16le -vn -ac 1 -ar 16k -map_metadata -1 test.wav

audio_file_path = "test-es-co.wav"
speech_sync_recognize(audio_file_path: audio_file_path, language_code: "es-CO", profanity_filter: true)

speech_sync_recognize(audio_file_path: audio_file_path, language_code: "es-CO", profanity_filter: true)
# Transcription: probando 1 2 3 probando 1 2 3 probando 1 2 3 probando Hola mundo Hola mundo Hola mundo Hola mundo
# => [<Google::Cloud::Speech::V1::SpeechRecognitionAlternative: transcript: "probando 1 2 3 probando 1 2 3 probando 1 2 3 probando Hola mundo Hola mundo Hola mundo Hola mundo", confidence: 0.9666613340377808, words: []>]