Skip to content

Commit

Permalink
Merge branch 'master' into fix/spellcheck
Browse files Browse the repository at this point in the history 
# Conflicts:
#	edenai_apis/apis/cohere/cohere_api.py
  • Loading branch information
@coscialp
coscialp committed Nov 15, 2023
2 parents a49f0a3 + 68efc93 commit 5132879
Show file tree
Hide file tree
Showing 11 changed files with 486 additions and 156 deletions.
4 changes: 3 additions & 1 deletion AVAILABLES_FEATURES_AND_PROVIDERS.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -237,7 +237,8 @@
| | lovoai |
| | microsoft |
| | openai |
| **text_to_speech_async** | lovoai |
| **text_to_speech_async** | amazon |
| | lovoai |

</details>
<details><summary>translation</summary>
Expand Down Expand Up @@ -319,6 +320,7 @@
|----------|-------------|
| **audio** | speech_to_text_async |
| | text_to_speech |
| | text_to_speech_async |
| **image** | explicit_content |
| | face_compare |
| | face_detection |
Expand Down
84 changes: 81 additions & 3 deletions edenai_apis/apis/amazon/amazon_audio_api.py
View file
Original file line number Diff line number Diff line change
@@ -1,9 +1,10 @@
import json
from pathlib import Path
from typing import Optional
from typing import Optional, Literal
import urllib
import uuid
import base64
import requests
from io import BufferedReader, BytesIO
from botocore.exceptions import BotoCoreError, ClientError
from edenai_apis.apis.amazon.helpers import (
Expand Down Expand Up @@ -36,9 +37,13 @@
AsyncResponseType,
ResponseType,
)
from edenai_apis.utils.upload_s3 import upload_file_bytes_to_s3, USER_PROCESS
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
from .config import audio_voices_ids, storage_clients
from edenai_apis.features.audio import TextToSpeechAsyncDataClass
from edenai_apis.loaders.data_loader import ProviderDataEnum
from edenai_apis.loaders.loaders import load_provider


class AmazonAudioApi(AudioInterface):
Expand Down Expand Up @@ -338,3 +343,76 @@ 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 = ""
) -> 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"]}

text = generate_right_ssml_text(
text, speaking_rate, speaking_pitch, speaking_volume
)

ext, audio_format, sampling = get_right_audio_support_and_sampling_rate(
audio_format, sampling_rate
)

params_update = {"OutputFormat": audio_format, "Text": text}
if sampling:
params_update["SampleRate"] = str(sampling)

params.update({**params_update})

if is_ssml(text):
params["TextType"] = "ssml"

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"))
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
) -> AsyncBaseResponseType[TextToSpeechAsyncDataClass]:
params = {"TaskId": provider_job_id}
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"))
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)
synthesis_task["OutputUri"] = file_url
response_file = requests.get(file_url)
print(response_file.content)
audio_content = BytesIO(response_file.content)
audio = base64.b64encode(audio_content.read()).decode("utf-8")
standardized_response = TextToSpeechAsyncDataClass(
audio_resource_url=file_url, audio=audio, voice_type=1
)
return AsyncResponseType[TextToSpeechAsyncDataClass](
original_response=synthesis_task,
standardized_response=standardized_response,
provider_job_id=provider_job_id
)
8 changes: 7 additions & 1 deletion edenai_apis/apis/amazon/config.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@ def clients(api_settings: Dict) -> Dict:
),
"texttospeech": boto3.client(
"polly",
region_name=api_settings["region_name"],
region_name=api_settings["ressource_region"],
aws_access_key_id=api_settings["aws_access_key_id"],
aws_secret_access_key=api_settings["aws_secret_access_key"],
),
Expand Down Expand Up @@ -89,6 +89,12 @@ def storage_clients(api_settings: Dict) -> Dict:
aws_access_key_id=api_settings["aws_access_key_id"],
aws_secret_access_key=api_settings["aws_secret_access_key"],
),
"texttospeech": boto3.resource(
"s3",
region_name=api_settings["ressource_region"],
aws_access_key_id=api_settings["aws_access_key_id"],
aws_secret_access_key=api_settings["aws_secret_access_key"],
)
}


Expand Down
159 changes: 158 additions & 1 deletion edenai_apis/apis/amazon/info.json
View file
Original file line number Diff line number Diff line change
Expand Up @@ -208,6 +208,163 @@
}
},
"version": "boto3 (v1.15.18)"
},
"text_to_speech_async": {
"constraints": {
"languages": [
"arb",
"en-US",
"fr-FR",
"es-ES",
"de-DE",
"en-GB",
"nl-NL",
"it-IT",
"ja-JP",
"cmn-CN",
"ru-RU",
"pt-BR",
"da-DK",
"ko-KR",
"pt-PT",
"en-AU",
"en-IN",
"hi-IN",
"is-IS",
"nb-NO",
"pl-PL",
"ro-RO",
"es-MX",
"es-US",
"sv-SE",
"tr-TR",
"cy-GB"
],
"audio_format": [
"mp3",
"ogg",
"pcm"
],
"voice_ids": {
"MALE": [
"da-DK_Mads_Standard",
"nl-NL_Ruben_Standard",
"en-AU_Russell_Neural",
"en-GB_Brian_Standard",
"en-GB_Brian_Neural",
"en-GB_Arthur_Neural",
"en-US_Joey_Standard",
"en-US_Joey_Neural",
"en-US_Justin_Standard",
"en-US_Justin_Neural",
"en-US_Kevin_Neural",
"en-US_Matthew_Standard",
"en-US_Matthew_Neural",
"en-US_Stephen_Neural",
"fr-FR_Mathieu_Standard",
"fr-FR_Remi_Neural",
"fr-CA_Liam_Neural",
"de-DE_Hans_Standard",
"is-IS_Karl_Standard",
"de-DE_Daniel_Neural",
"is-IS_Karl_Neural",
"it-IT_Giorgio_Standard",
"it-IT_Adriano_Neural",
"ja-JP_Takumi_Standard",
"ja-JP_Takumi_Neural",
"pl-PL_Jacek_Standard",
"pl-PL_Jan_Standard",
"pt-BR_Ricardo_Standard",
"pt-BR_Thiago_Neural",
"pt-PT_Cristiano_Standard",
"ru-RU_Maxim_Standard",
"es-ES_Enrique_Standard",
"es-ES_Sergio_Neural",
"es-MX_Andres_Neural",
"es-US_Miguel_Standard",
"es-US_Pedro_Neural"
],
"FEMALE": [
"arb_Zeina_Standard",
"ar-AE_Hala_Neural",
"ca-ES_Arlet_Neural",
"yue-CN_Hiujin_Neural",
"cmn-CN_Zhiyu_Standard",
"cmn-CN_Zhiyu_Neural",
"da-DK_Naja_Standard",
"nl-NL_Laura_Neural",
"nl-NL_Lotte_Standard",
"en-AU_Nicole_Neural",
"en-AU_Olivia_Standard",
"en-GB_Amy_Standard",
"en-GB_Amy_Neural",
"en-GB_Emma_Standard",
"en-GB_Emma_Neural",
"en-IN_Aditi_Standard",
"en-IN_Raveena_Standard",
"en-IN_Kajal_Neural",
"en-NZ_Aria_Neural",
"en-ZA_Ayanda_Neural",
"en-US_Ivy_Standard",
"en-US_Ivy_Neural",
"en-US_Joanna_Standard",
"en-US_Joanna_Neural",
"en-US_Kendra_Standard",
"en-US_Kendra_Neural",
"en-US_Kimberly_Standard",
"en-US_Kimberly_Neural",
"en-US_Salli_Standard",
"en-US_Salli_Neural",
"en-US_Ruth_Neural",
"fi-FI_Suvi_Neural",
"fr-FR_Celine_Standard",
"fr-FR_Lea_Standard",
"fr-FR_Lea_Neural",
"fr-CA_Chantal_Standard",
"fr-CA_Gabrielle_Neural",
"de-DE_Marlene_Standard",
"de-DE_Vicki_Standard",
"de-DE_Vicki_Neural",
"de-AT_Hannah_Neural",
"hi-IN_Aditi_Standard",
"hi-IN_Kajal_Neural",
"is-IS_Dora_Standard",
"it-IT_Carla_Standard",
"it-IT_Bianca_Standard",
"it-IT_Bianca_Neural",
"ja-JP_Mizuki_Standard",
"ja-JP_Kazuha_Neural",
"ja-JP_Tomoko_Neural",
"ko-KR_Seoyeon_Standard",
"ko-KR_Seoyeon_Neural",
"nb-NO_Liv_Standard",
"pl-PL_Ewa_Standard",
"pl-PL_Maja_Standard",
"pl-PL_Ola_Neural",
"pt-BR_Camila_Standard",
"pt-BR_Camila_Neural",
"pt-BR_Vitoria_Standard",
"pt-BR_Vitoria_Neural",
"pt-PT_Ines_Standard",
"pt-PT_Ines_Neural",
"ro-RO_Carmen_Standard",
"ru-RU_Tatyana_Standard",
"es-ES_Conchita_Standard",
"es-ES_Lucia_Standard",
"es-ES_Lucia_Neural",
"es-MX_Mia_Standard",
"es-MX_Mia_Neural",
"es-US_Lupe_Standard",
"es-US_Lupe_Neural",
"es-US_Penelope_Standard",
"sv-SE_Astrid_Standard",
"sv-SE_Elin_Neural",
"tr-TR_Filiz_Standard",
"cy-GB_Gwyneth_Standard"
]
}
},
"version": "boto3 (v1.15.18)"
}
},
"ocr": {
Expand Down Expand Up @@ -538,7 +695,7 @@
"version": "boto3 (v1.15.18)"
},
"object_detection": {
"constraints":{
"constraints": {
},
"version": "boto3 (v1.15.18)"
},
Expand Down
Loading

0 comments on commit 5132879

Please sign in to comment.