or run

npx @tessl/cli init
Log in

Version

Tile

Overview

Evals

Files

Files

docs

chat-interface.mdclient-management.mddocument-prompt-template.mdembeddings.mdevaluation.mdexplanations.mdindex.mdprompt-construction.mdsteering.mdstructured-output.mdtext-completion.mdtokenization.mdtranslation.mdutilities.md

translation.mddocs/

0

# Translation

1

2

Translate text between languages using Aleph Alpha's translation models. Provides high-quality translations with quality scoring and segment-level analysis.

3

4

## Capabilities

5

6

### Translation Requests

7

8

Configure translation operations with source text and target language specification.

9

10

```python { .api }

11

class TranslationRequest:

12

model: str

13

source: str

14

target_language: str

15

16

def __init__(self, model: str, source: str, target_language: str):

17

"""

18

Request for translation.

19

20

Parameters:

21

- model: The name of the model to be used for the translation

22

- source: The input text to be translated

23

- target_language: The desired target language using ISO 639 language codes

24

(e.g., "en" for English, "de" for German, "fr" for French)

25

"""

26

27

def to_json(self) -> dict:

28

"""Convert the request to a JSON-serializable dictionary."""

29

```

30

31

### Translation Responses

32

33

Structured response containing translated text with quality metrics and optional segment analysis.

34

35

```python { .api }

36

class TranslationResponse:

37

translation: str

38

score: float

39

segments: Optional[List[TranslationSegment]]

40

num_tokens_prompt_total: int

41

num_tokens_generated: int

42

43

def __init__(

44

self,

45

translation: str,

46

score: float,

47

segments: Optional[List[TranslationSegment]],

48

num_tokens_prompt_total: int,

49

num_tokens_generated: int

50

):

51

"""

52

Response from a translation request.

53

54

Attributes:

55

- translation: The complete translated output text

56

- score: Overall quality estimate on a scale of 0 to 1

57

- segments: List of translated segments (may be None)

58

- num_tokens_prompt_total: Total tokens in prompt (may be zero)

59

- num_tokens_generated: Total tokens generated (may be zero)

60

"""

61

62

@classmethod

63

def from_json(cls, json: dict) -> TranslationResponse:

64

"""Create a TranslationResponse from a JSON dictionary."""

65

```

66

67

### Translation Segments

68

69

Individual segments of translated text with quality scoring.

70

71

```python { .api }

72

class TranslationSegment:

73

source: str

74

translation: str

75

score: float

76

77

def __init__(self, source: str, translation: str, score: float):

78

"""

79

A segment of translated text with its source and quality score.

80

81

Parameters:

82

- source: The input text to be translated

83

- translation: The translated output text of the segment

84

- score: Quality estimate for this segment on a scale of 0 to 1

85

"""

86

```

87

88

### Translation Methods

89

90

Translate text using synchronous and asynchronous clients.

91

92

```python { .api }

93

def translate(self, request: TranslationRequest) -> TranslationResponse:

94

"""

95

Translate text to target language.

96

97

Parameters:

98

- request: Translation configuration

99

100

Returns:

101

TranslationResponse with translated text and quality metrics

102

"""

103

104

async def translate(self, request: TranslationRequest) -> TranslationResponse:

105

"""

106

Translate text to target language (async).

107

108

Parameters:

109

- request: Translation configuration

110

111

Returns:

112

TranslationResponse with translated text and quality metrics

113

"""

114

```

115

116

### Usage Examples

117

118

Basic translation with quality assessment:

119

120

```python

121

from aleph_alpha_client import Client, TranslationRequest

122

123

client = Client(token="your-api-token")

124

125

# Simple translation

126

request = TranslationRequest(

127

model="luminous-extended",

128

source="Hello, how are you today?",

129

target_language="de" # German

130

)

131

132

response = client.translate(request)

133

print(f"Translation: {response.translation}")

134

print(f"Quality Score: {response.score:.2f}")

135

136

# Translation with segment analysis

137

request = TranslationRequest(

138

model="luminous-extended",

139

source="Good morning! I hope you're having a wonderful day. The weather is beautiful today.",

140

target_language="fr" # French

141

)

142

143

response = client.translate(request)

144

print(f"Full Translation: {response.translation}")

145

print(f"Overall Quality: {response.score:.2f}")

146

147

if response.segments:

148

print("\nSegment Analysis:")

149

for i, segment in enumerate(response.segments):

150

print(f"Segment {i+1}:")

151

print(f" Source: {segment.source}")

152

print(f" Translation: {segment.translation}")

153

print(f" Quality: {segment.score:.2f}")

154

155

# Async translation

156

import asyncio

157

158

async def translate_async():

159

async with AsyncClient(token="your-api-token") as client:

160

request = TranslationRequest(

161

model="luminous-extended",

162

source="Artificial intelligence is transforming our world.",

163

target_language="es" # Spanish

164

)

165

166

response = await client.translate(request)

167

print(f"Async Translation: {response.translation}")

168

return response

169

170

# Run async translation

171

response = asyncio.run(translate_async())

172

173

# Multiple language translation

174

languages = {

175

"de": "German",

176

"fr": "French",

177

"es": "Spanish",

178

"it": "Italian"

179

}

180

181

source_text = "Technology connects people across the globe."

182

183

for lang_code, lang_name in languages.items():

184

request = TranslationRequest(

185

model="luminous-extended",

186

source=source_text,

187

target_language=lang_code

188

)

189

190

response = client.translate(request)

191

print(f"{lang_name}: {response.translation} (Quality: {response.score:.2f})")

192

```