apunkt/MetaGPT
1
0
Fork 0
mirror of https://github.com/FoundationAgents/MetaGPT.git synced 2026-05-15 11:02:36 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

feat: +iflytek tts

Browse source
This commit is contained in:
莘权 马 2023-09-05 19:10:16 +08:00
parent 5c627df6c4
commit c1aa932210
3 changed files with 85 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

29
metagpt/learn/text_to_speech.py
View file

@ -11,6 +11,7 @@ import openai
from metagpt.config import CONFIG
from metagpt.const import BASE64_FORMAT
from metagpt.tools.azure_tts import oas3_azsure_tts
from metagpt.tools.iflytek_tts import oas3_iflytek_tts
from metagpt.utils.s3 import S3
@ -22,6 +23,9 @@ async def text_to_speech(
role="Girl",
subscription_key="",
region="",
iflytek_app_id="",
iflytek_api_key="",
iflytek_api_secret="",
**kwargs,
):
"""Text to speech
@ -34,16 +38,35 @@ async def text_to_speech(
:param text: The text used for voice conversion.
:param subscription_key: key is used to access your Azure AI service API, see: `https://portal.azure.com/` > `Resource Management` > `Keys and Endpoint`
:param region: This is the location (or region) of your resource. You may need to use this field when making calls to this API.
:return: Returns the Base64-encoded .wav file data if successful, otherwise an empty string.
:param iflytek_app_id: Application ID is used to access your iFlyTek service API, see: `https://console.xfyun.cn/services/tts`
:param iflytek_api_key: WebAPI argument, see: `https://console.xfyun.cn/services/tts`
:param iflytek_api_secret: WebAPI argument, see: `https://console.xfyun.cn/services/tts`
:return: Returns the Base64-encoded .wav/.mp3 file data if successful, otherwise an empty string.
"""
audio_declaration = "data:audio/wav;base64,"
if (CONFIG.AZURE_TTS_SUBSCRIPTION_KEY and CONFIG.AZURE_TTS_REGION) or (subscription_key and region):
audio_declaration = "data:audio/wav;base64,"
base64_data = await oas3_azsure_tts(text, lang, voice, style, role, subscription_key, region)
s3 = S3()
url = await s3.cache(data=base64_data, file_ext=".wav", format=BASE64_FORMAT)
if url:
return f"[{text}]({url})"
return audio_declaration + base64_data if base64_data else base64_data
if (CONFIG.IFLYTEK_APP_ID and CONFIG.IFLYTEK_API_KEY and CONFIG.IFLYTEK_API_SECRET) or (
iflytek_app_id and iflytek_api_key and iflytek_api_secret
):
audio_declaration = "data:audio/mp3;base64,"
base64_data = await oas3_iflytek_tts(
text=text, app_id=iflytek_app_id, api_key=iflytek_api_key, api_secret=iflytek_api_secret
)
s3 = S3()
url = await s3.cache(data=base64_data, file_ext=".mp3", format=BASE64_FORMAT)
if url:
return f"[{text}]({url})"
return audio_declaration + base64_data if base64_data else base64_data
raise openai.error.InvalidRequestError(message="AZURE_TTS_SUBSCRIPTION_KEY and AZURE_TTS_REGION error", param={})
raise openai.error.InvalidRequestError(
message="AZURE_TTS_SUBSCRIPTION_KEY, AZURE_TTS_REGION, IFLYTEK_APP_ID, IFLYTEK_API_KEY, IFLYTEK_API_SECRET error",
param={},
)