File size: 3,104 Bytes
ab2ded1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
#
#  Copyright 2024 The InfiniFlow Authors. All Rights Reserved.
#
#  Licensed under the Apache License, Version 2.0 (the "License");
#  you may not use this file except in compliance with the License.
#  You may obtain a copy of the License at
#
#      http://www.apache.org/licenses/LICENSE-2.0
#
#  Unless required by applicable law or agreed to in writing, software
#  distributed under the License is distributed on an "AS IS" BASIS,
#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
#  See the License for the specific language governing permissions and
#  limitations under the License.
#
from openai.lib.azure import AzureOpenAI
from zhipuai import ZhipuAI
import io
from abc import ABC
from ollama import Client
from openai import OpenAI
import os
import json
from rag.utils import num_tokens_from_string


class Base(ABC):
    def __init__(self, key, model_name):
        pass

    def transcription(self, audio, **kwargs):
        transcription = self.client.audio.transcriptions.create(
            model=self.model_name,
            file=audio,
            response_format="text"
        )
        return transcription.text.strip(), num_tokens_from_string(transcription.text.strip())


class GPTSeq2txt(Base):
    def __init__(self, key, model_name="whisper-1", base_url="https://api.openai.com/v1"):
        if not base_url: base_url = "https://api.openai.com/v1"
        self.client = OpenAI(api_key=key, base_url=base_url)
        self.model_name = model_name


class QWenSeq2txt(Base):
    def __init__(self, key, model_name="paraformer-realtime-8k-v1", **kwargs):
        import dashscope
        dashscope.api_key = key
        self.model_name = model_name

    def transcription(self, audio, format):
        from http import HTTPStatus
        from dashscope.audio.asr import Recognition

        recognition = Recognition(model=self.model_name,
                                  format=format,
                                  sample_rate=16000,
                                  callback=None)
        result = recognition.call(audio)

        ans = ""
        if result.status_code == HTTPStatus.OK:
            for sentence in result.get_sentence():
                ans += str(sentence + '\n')
            return ans, num_tokens_from_string(ans)

        return "**ERROR**: " + result.message, 0


class OllamaSeq2txt(Base):
    def __init__(self, key, model_name, lang="Chinese", **kwargs):
        self.client = Client(host=kwargs["base_url"])
        self.model_name = model_name
        self.lang = lang


class AzureSeq2txt(Base):
    def __init__(self, key, model_name, lang="Chinese", **kwargs):
        self.client = AzureOpenAI(api_key=key, azure_endpoint=kwargs["base_url"], api_version="2024-02-01")
        self.model_name = model_name
        self.lang = lang


class XinferenceSeq2txt(Base):
    def __init__(self, key, model_name="", base_url=""):
        self.client = OpenAI(api_key="xxx", base_url=base_url)
        self.model_name = model_name