Google Cloud Speech Translate Operators¶
GcpTranslateSpeechOperator¶
Recognizes speech in audio input and translates it.
For parameter definition, take a look at
airflow.contrib.operators.gcp_translate_speech_operator.GcpTranslateSpeechOperator
Arguments¶
Config and audio arguments need to be dicts or objects of corresponding classes from
google.cloud.speech_v1.types
module.
for more information, see: https://googleapis.github.io/google-cloud-python/latest/speech/gapic/v1/api.html#google.cloud.speech_v1.SpeechClient.recognize
CONFIG = {"encoding": "LINEAR16", "language_code": "en_US"}
AUDIO = {"uri": "gs://{bucket}/{object}".format(bucket=BUCKET_NAME, object=FILENAME)}
Arguments for translation need to be specified.
TARGET_LANGUAGE = 'pl'
FORMAT = 'text'
MODEL = 'base'
SOURCE_LANGUAGE = None
Using the operator¶
translate_speech_task = GcpTranslateSpeechOperator(
project_id=GCP_PROJECT_ID,
audio=AUDIO,
config=CONFIG,
target_language=TARGET_LANGUAGE,
format_=FORMAT,
source_language=SOURCE_LANGUAGE,
model=MODEL,
task_id='translate_speech_task'
)
Templating¶
template_fields = ('target_language', 'format_', 'source_language', 'model', 'project_id', 'gcp_conn_id')