You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
"description": "Whisper is a pre-trained model for automatic speech recognition (ASR) and speech translation. ",
6
+
"task": {
7
+
"id": "dfce1c48-2a81-462e-a7fd-de97ce985207",
8
+
"name": "Automatic Speech Recognition",
9
+
"description": "Automatic speech recognition (ASR) models convert a speech signal, typically an audio input, to text."
10
+
},
11
+
"tags": [],
12
+
"properties": [
13
+
{
14
+
"property_id": "beta",
15
+
"value": "true"
16
+
}
17
+
],
18
+
"schema": {
19
+
"input": {
20
+
"type": "object",
21
+
"properties": {
22
+
"audio": {
23
+
"type": "string",
24
+
"description": "Base64 encoded value of the audio data."
25
+
},
26
+
"task": {
27
+
"type": "string",
28
+
"default": "transcribe",
29
+
"description": "Supported tasks are 'translate' or 'transcribe'."
30
+
},
31
+
"language": {
32
+
"type": "string",
33
+
"default": "en",
34
+
"description": "The language of the audio being transcribed or translated."
35
+
},
36
+
"vad_filter": {
37
+
"type": "string",
38
+
"default": "false",
39
+
"description": "Preprocess the audio with a voice activity detection model."
40
+
},
41
+
"initial_prompt": {
42
+
"type": "string",
43
+
"description": "A text prompt to help provide context to the model on the contents of the audio."
44
+
},
45
+
"prefix": {
46
+
"type": "string",
47
+
"description": "The prefix it appended the the beginning of the output of the transcription and can guide the transcription result."
48
+
}
49
+
},
50
+
"required": [
51
+
"audio"
52
+
]
53
+
},
54
+
"output": {
55
+
"type": "object",
56
+
"contentType": "application/json",
57
+
"properties": {
58
+
"transcription_info": {
59
+
"type": "object",
60
+
"properties": {
61
+
"language": {
62
+
"type": "string",
63
+
"description": "The language of the audio being transcribed or translated."
64
+
},
65
+
"language_probability": {
66
+
"type": "number",
67
+
"description": "The confidence level or probability of the detected language being accurate, represented as a decimal between 0 and 1."
68
+
},
69
+
"duration": {
70
+
"type": "number",
71
+
"description": "The total duration of the original audio file, in seconds."
72
+
},
73
+
"duration_after_vad": {
74
+
"type": "number",
75
+
"description": "The duration of the audio after applying Voice Activity Detection (VAD) to remove silent or irrelevant sections, in seconds."
76
+
}
77
+
}
78
+
},
79
+
"text": {
80
+
"type": "string",
81
+
"description": "The complete transcription of the audio."
82
+
},
83
+
"word_count": {
84
+
"type": "number",
85
+
"description": "The total number of words in the transcription."
86
+
},
87
+
"segments": {
88
+
"type": "object",
89
+
"properties": {
90
+
"start": {
91
+
"type": "number",
92
+
"description": "The starting time of the segment within the audio, in seconds."
93
+
},
94
+
"end": {
95
+
"type": "number",
96
+
"description": "The ending time of the segment within the audio, in seconds."
97
+
},
98
+
"text": {
99
+
"type": "string",
100
+
"description": "The transcription of the segment."
101
+
},
102
+
"temperature": {
103
+
"type": "number",
104
+
"description": "The temperature used in the decoding process, controlling randomness in predictions. Lower values result in more deterministic outputs."
105
+
},
106
+
"avg_logprob": {
107
+
"type": "number",
108
+
"description": "The average log probability of the predictions for the words in this segment, indicating overall confidence."
109
+
},
110
+
"compression_ratio": {
111
+
"type": "number",
112
+
"description": "The compression ratio of the input to the output, measuring how much the text was compressed during the transcription process."
113
+
},
114
+
"no_speech_prob": {
115
+
"type": "number",
116
+
"description": "The probability that the segment contains no speech, represented as a decimal between 0 and 1."
117
+
},
118
+
"words": {
119
+
"type": "array",
120
+
"items": {
121
+
"type": "object",
122
+
"properties": {
123
+
"word": {
124
+
"type": "string",
125
+
"description": "The individual word transcribed from the audio."
126
+
},
127
+
"start": {
128
+
"type": "number",
129
+
"description": "The starting time of the word within the audio, in seconds."
130
+
},
131
+
"end": {
132
+
"type": "number",
133
+
"description": "The ending time of the word within the audio, in seconds."
134
+
}
135
+
}
136
+
}
137
+
}
138
+
}
139
+
},
140
+
"vtt": {
141
+
"type": "string",
142
+
"description": "The transcription in WebVTT format, which includes timing and text information for use in subtitles."
0 commit comments