From a2aab8df7e818e5180c0a22e1272f682c5f899a8 Mon Sep 17 00:00:00 2001 From: CaptainB Date: Fri, 20 Sep 2024 10:58:33 +0800 Subject: [PATCH] =?UTF-8?q?fix:=20=E5=A4=84=E7=90=86opanai=E8=AF=AD?= =?UTF-8?q?=E9=9F=B3=E8=AF=86=E5=88=AB=E6=8A=A5=E9=94=99=E7=9A=84=E9=97=AE?= =?UTF-8?q?=E9=A2=98?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --bug=1046791 --user=刘瑞斌 【应用】使用 openai 语音输入模型、输出模型进行输入输出报错 https://www.tapd.cn/57709429/s/1582133 --- .../impl/openai_model_provider/model/stt.py | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/apps/setting/models_provider/impl/openai_model_provider/model/stt.py b/apps/setting/models_provider/impl/openai_model_provider/model/stt.py index 66b2daeda..0b5f9a4b9 100644 --- a/apps/setting/models_provider/impl/openai_model_provider/model/stt.py +++ b/apps/setting/models_provider/impl/openai_model_provider/model/stt.py @@ -1,3 +1,5 @@ +import asyncio +import io from typing import Dict from openai import OpenAI @@ -49,5 +51,9 @@ class OpenAISpeechToText(MaxKBBaseModel, BaseSpeechToText): base_url=self.api_base, api_key=self.api_key ) - res = client.audio.transcriptions.create(model=self.model, language="zh", file=audio_file) + audio_data = audio_file.read() + buffer = io.BytesIO(audio_data) + buffer.name = "file.mp3" # this is the important line + res = client.audio.transcriptions.create(model=self.model, language="zh", file=buffer) return res.text +