ExternalVendorCode/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-06-18 14:58:09 +00:00
Code Issues Actions 4 Packages Projects Releases Wiki Activity

go : adding features to the go-whisper example, go ci, etc (#384)

Browse Source 
* Updated bindings so they can be used in third pary packages.

* Updated makefiles to set FMA flag on optionally, for xeon E5 on Darwin

* Added test script

* Changes for examples

* Reverted

* Made the NewContext method private
This commit is contained in:
David Thorpe
2023-01-07 19:21:43 +00:00
committed by GitHub
parent f30b5d322c
commit f078a6f20e
10 changed files with 370 additions and 31 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
bindings/go/pkg/whisper/interface.go
View File

@ -20,6 +20,9 @@ type Model interface {
// 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
}
@ -27,8 +30,18 @@ type Model interface {
// Context is the speach recognition context.
type Context interface {
SetLanguage(string) error // Set the language to use for speech recognition.
SetTranslate(bool) // Set translate flag
IsMultilingual() bool // Return true if the model is multilingual.
Language() string // Get language
SetSpeedup(bool) // Set speedup flag
SetOffset(time.Duration) // Set offset
SetDuration(time.Duration) // Set duration
SetThreads(uint) // Set number of threads to use
SetSpeedup(bool) // Set speedup flag
SetTokenThreshold(float32) // Set timestamp token probability threshold
SetTokenSumThreshold(float32) // Set timestamp token sum probability threshold
SetMaxSegmentLength(uint) // Set max segment length in characters
SetMaxTokensPerSegment(uint) // Set max tokens per segment (0 = no limit)
// Process mono audio data and return any errors.
// If defined, newly generated segments are passed to the
@ -38,6 +51,15 @@ type Context interface {
// 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
}
// Segment is the text result of a speech recognition.