Spaces:
Running
Running
File size: 3,827 Bytes
5e456c0 19d7f69 6eea7b7 5e456c0 728fcbe 5e456c0 6eea7b7 5e456c0 ccaa693 728fcbe 5e456c0 1830e27 728fcbe 5b36f0b 5e456c0 6eea7b7 5e456c0 728fcbe 89e8314 4a56623 89e8314 4a56623 5e456c0 38f9f3b 5e456c0 |
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 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 |
package whisper
import (
"io"
"time"
)
///////////////////////////////////////////////////////////////////////////////
// TYPES
// SegmentCallback is the callback function for processing segments in real
// time. It is called during the Process function
type SegmentCallback func(Segment)
// ProgressCallback is the callback function for reporting progress during
// processing. It is called during the Process function
type ProgressCallback func(int)
// EncoderBeginCallback is the callback function for checking if we want to
// continue processing. It is called during the Process function
type EncoderBeginCallback func() bool
// Model is the interface to a whisper model. Create a new model with the
// function whisper.New(string)
type Model interface {
io.Closer
// Return a new speech-to-text context.
NewContext() (Context, error)
// Return true if the model is multilingual.
IsMultilingual() bool
// Return all languages supported.
Languages() []string
}
// Context is the speech recognition context.
type Context interface {
SetLanguage(string) error // Set the language to use for speech recognition, use "auto" for auto detect language.
SetTranslate(bool) // Set translate flag
IsMultilingual() bool // Return true if the model is multilingual.
Language() string // Get language
DetectedLanguage() string // Get detected language
SetOffset(time.Duration) // Set offset
SetDuration(time.Duration) // Set duration
SetThreads(uint) // Set number of threads to use
SetSplitOnWord(bool) // Set split on word flag
SetTokenThreshold(float32) // Set timestamp token probability threshold
SetTokenSumThreshold(float32) // Set timestamp token sum probability threshold
SetMaxSegmentLength(uint) // Set max segment length in characters
SetTokenTimestamps(bool) // Set token timestamps flag
SetMaxTokensPerSegment(uint) // Set max tokens per segment (0 = no limit)
SetAudioCtx(uint) // Set audio encoder context
SetMaxContext(n int) // Set maximum number of text context tokens to store
SetBeamSize(n int) // Set Beam Size
SetEntropyThold(t float32) // Set Entropy threshold
SetInitialPrompt(prompt string) // Set initial prompt
SetTemperature(t float32) // Set temperature
SetTemperatureFallback(t float32) // Set temperature incrementation
// Process mono audio data and return any errors.
// If defined, newly generated segments are passed to the
// callback function during processing.
Process([]float32, EncoderBeginCallback, SegmentCallback, ProgressCallback) error
// After process is called, return segments until the end of the stream
// is reached, when io.EOF is returned.
NextSegment() (Segment, error)
IsBEG(Token) bool // Test for "begin" token
IsSOT(Token) bool // Test for "start of transcription" token
IsEOT(Token) bool // Test for "end of transcription" token
IsPREV(Token) bool // Test for "start of prev" token
IsSOLM(Token) bool // Test for "start of lm" token
IsNOT(Token) bool // Test for "No timestamps" token
IsLANG(Token, string) bool // Test for token associated with a specific language
IsText(Token) bool // Test for text token
// Timings
PrintTimings()
ResetTimings()
SystemInfo() string
}
// Segment is the text result of a speech recognition.
type Segment struct {
// Segment Number
Num int
// Time beginning and end timestamps for the segment.
Start, End time.Duration
// The text of the segment.
Text string
// The tokens of the segment.
Tokens []Token
}
// Token is a text or special token
type Token struct {
Id int
Text string
P float32
Start, End time.Duration
}
|