Skip to content

Commit

Permalink
[fix] amazon NoneType object has no attribute get
Browse files Browse the repository at this point in the history
  • Loading branch information
@DninoAdnane
DninoAdnane committed Dec 8, 2023
1 parent 79a8347 commit 4621933
Showing 1 changed file with 69 additions and 43 deletions.
112 changes: 69 additions & 43 deletions edenai_apis/apis/amazon/amazon_audio_api.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -33,8 +33,14 @@
AsyncResponseType,
ResponseType,
)
from edenai_apis.utils.upload_s3 import upload_file_bytes_to_s3, USER_PROCESS, get_s3_file_url, URL_LONG_PERIOD, \
get_cloud_front_file_url, s3_client_load
from edenai_apis.utils.upload_s3 import (
upload_file_bytes_to_s3,
USER_PROCESS,
get_s3_file_url,
URL_LONG_PERIOD,
get_cloud_front_file_url,
s3_client_load,
)

from .config import audio_voices_ids, storage_clients
from edenai_apis.features.audio import TextToSpeechAsyncDataClass
Expand Down Expand Up @@ -77,8 +83,10 @@ def audio__text_to_speech(
if is_ssml(text):
params["TextType"] = "ssml"

response = handle_amazon_call(self.clients["texttospeech"].synthesize_speech, **params)

response = handle_amazon_call(
self.clients["texttospeech"].synthesize_speech, **params
)

audio_content = BytesIO(response["AudioStream"].read())

# convert 'StreamBody' to b64
Expand Down Expand Up @@ -116,11 +124,13 @@ def _create_vocabulary(self, language: str, list_vocabs: list):
list_vocabs = ["-".join(vocab.strip().split()) for vocab in list_vocabs]
vocab_name = str(uuid.uuid4())
payload = {
"LanguageCode" : language,
"VocabularyName" : vocab_name,
"Phrases": list_vocabs
"LanguageCode": language,
"VocabularyName": vocab_name,
"Phrases": list_vocabs,
}
response = handle_amazon_call(self.clients["speech"].create_vocabulary, **payload)
response = handle_amazon_call(
self.clients["speech"].create_vocabulary, **payload
)

return vocab_name

Expand All @@ -133,7 +143,7 @@ def _launch_transcribe(
vocab_name: Optional[str] = None,
initiate_vocab: bool = False,
format: str = "wav",
provider_params: dict = dict()
provider_params: dict = dict(),
):
if speakers < 2:
speakers = 2
Expand Down Expand Up @@ -169,11 +179,8 @@ def _launch_transcribe(
raise ProviderException(str(exc)) from exc

def _delete_vocabularies(self, vocab_name):
payload = {
"VocabularyName" : vocab_name
}
payload = {"VocabularyName": vocab_name}
handle_amazon_call(self.clients["speech"].delete_vocabulary, **payload)


def audio__speech_to_text_async__launch_job(
self,
Expand All @@ -185,7 +192,7 @@ def audio__speech_to_text_async__launch_job(
audio_attributes: tuple,
model: str = None,
file_url: str = "",
provider_params = dict()
provider_params=dict(),
) -> AsyncLaunchJobResponseType:
export_format, channels, frame_rate = audio_attributes

Expand All @@ -196,7 +203,7 @@ def audio__speech_to_text_async__launch_job(
if language is None:
raise ProviderException(
"Cannot launch with vocabulary when language is auto-detect.",
code = 400
code=400,
)
vocab_name = self._create_vocabulary(language, vocabulary)
self._launch_transcribe(
Expand All @@ -207,7 +214,7 @@ def audio__speech_to_text_async__launch_job(
vocab_name,
True,
format=export_format,
provider_params=provider_params
provider_params=provider_params,
)
return AsyncLaunchJobResponseType(
provider_job_id=f"{filename}EdenAI{vocab_name}"
Expand Down Expand Up @@ -266,9 +273,11 @@ def audio__speech_to_text_async__get_job_result(
)

# check transcribe status
payload= {"TranscriptionJobName" : job_id}
job_details = handle_amazon_call(self.clients["speech"].get_transcription_job, **payload)

payload = {"TranscriptionJobName": job_id}
job_details = handle_amazon_call(
self.clients["speech"].get_transcription_job, **payload
)

job_status = job_details["TranscriptionJob"]["TranscriptionJobStatus"]
if job_status == "COMPLETED":
# delete vocabulary
Expand All @@ -286,10 +295,8 @@ def audio__speech_to_text_async__get_job_result(
diarization_entries = []
words_info = original_response["results"]["items"]
speakers = (
original_response["results"]
.get("speaker_labels", {})
.get("speakers", 0)
)
original_response["results"].get("speaker_labels", {}) or {}
).get("speakers", 0)

for word_info in words_info:
if word_info.get("speaker_label"):
Expand Down Expand Up @@ -339,23 +346,29 @@ def audio__speech_to_text_async__get_job_result(
return AsyncPendingResponseType[SpeechToTextAsyncDataClass](
provider_job_id=provider_job_id
)

def audio__text_to_speech_async__launch_job(
self,
language: str,
text: str,
option: str,
voice_id: str,
audio_format: str,
speaking_rate: int,
speaking_pitch: int,
speaking_volume: int,
sampling_rate: int,
file_url: str = ""
self,
language: str,
text: str,
option: str,
voice_id: str,
audio_format: str,
speaking_rate: int,
speaking_pitch: int,
speaking_volume: int,
sampling_rate: int,
file_url: str = "",
) -> AsyncLaunchJobResponseType:
_, voice_id_name, engine = voice_id.split("_")
engine = engine.lower()

params = {"Engine": engine, "VoiceId": voice_id_name, "OutputFormat": "mp3", "OutputS3BucketName": self.api_settings["users_resource_bucket"]}
params = {
"Engine": engine,
"VoiceId": voice_id_name,
"OutputFormat": "mp3",
"OutputS3BucketName": self.api_settings["users_resource_bucket"],
}

text = generate_right_ssml_text(
text, speaking_rate, speaking_pitch, speaking_volume
Expand All @@ -374,31 +387,44 @@ def audio__text_to_speech_async__launch_job(
if is_ssml(text):
params["TextType"] = "ssml"

response = handle_amazon_call(self.clients["texttospeech"].start_speech_synthesis_task, **params)
response = handle_amazon_call(
self.clients["texttospeech"].start_speech_synthesis_task, **params
)
synthesis_task = response["SynthesisTask"]
if synthesis_task["TaskStatus"] == "failed":
raise ProviderException(synthesis_task.get("TaskStatusReason", "Amazon returned a job status: failed"))
raise ProviderException(
synthesis_task.get(
"TaskStatusReason", "Amazon returned a job status: failed"
)
)
print(synthesis_task["TaskId"])
return AsyncLaunchJobResponseType(provider_job_id=synthesis_task["TaskId"])

def audio__text_to_speech_async__get_job_result(
self,
provider_job_id: str
self, provider_job_id: str
) -> AsyncBaseResponseType[TextToSpeechAsyncDataClass]:
params = {"TaskId": provider_job_id}
response = handle_amazon_call(self.clients["texttospeech"].get_speech_synthesis_task, **params)
response = handle_amazon_call(
self.clients["texttospeech"].get_speech_synthesis_task, **params
)
synthesis_task = response["SynthesisTask"]
status = synthesis_task["TaskStatus"]
if status == "failed":
raise ProviderException(synthesis_task.get("TaskStatusReason", "Amazon returned a job status: failed"))
raise ProviderException(
synthesis_task.get(
"TaskStatusReason", "Amazon returned a job status: failed"
)
)
elif status == "inProgress" or status == "scheduled":
return AsyncPendingResponseType[TextToSpeechAsyncDataClass](
provider_job_id=provider_job_id
)
else:
output_uri = synthesis_task.get("OutputUri", "")
s3_client_load()
file_url = get_cloud_front_file_url(output_uri.split('/')[-1], URL_LONG_PERIOD)
file_url = get_cloud_front_file_url(
output_uri.split("/")[-1], URL_LONG_PERIOD
)
synthesis_task["OutputUri"] = file_url
response_file = requests.get(file_url)
print(response_file.content)
Expand All @@ -410,5 +436,5 @@ def audio__text_to_speech_async__get_job_result(
return AsyncResponseType[TextToSpeechAsyncDataClass](
original_response=synthesis_task,
standardized_response=standardized_response,
provider_job_id=provider_job_id
provider_job_id=provider_job_id,
)

0 comments on commit 4621933

Please sign in to comment.