mirror of
https://github.com/mudler/LocalAI.git
synced 2024-12-24 23:06:42 +00:00
af9e5a2d05
* Revert "fix(fncall): fix regression introduced in #1963 (#2048)" This reverts commit6b06d4e0af
. * Revert "fix: action-tmate back to upstream, dead code removal (#2038)" This reverts commitfdec8a9d00
. * Revert "feat(grpc): return consumed token count and update response accordingly (#2035)" This reverts commite843d7df0e
. * Revert "refactor: backend/service split, channel-based llm flow (#1963)" This reverts commiteed5706994
. * feat(grpc): return consumed token count and update response accordingly Fixes: #1920 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
101 lines
2.0 KiB
Go
101 lines
2.0 KiB
Go
package main
|
|
|
|
import (
|
|
"fmt"
|
|
"os"
|
|
"os/exec"
|
|
"path/filepath"
|
|
|
|
"github.com/ggerganov/whisper.cpp/bindings/go/pkg/whisper"
|
|
"github.com/go-audio/wav"
|
|
"github.com/go-skynet/LocalAI/core/schema"
|
|
)
|
|
|
|
func runCommand(command []string) (string, error) {
|
|
cmd := exec.Command(command[0], command[1:]...)
|
|
cmd.Env = os.Environ()
|
|
out, err := cmd.CombinedOutput()
|
|
return string(out), err
|
|
}
|
|
|
|
// AudioToWav converts audio to wav for transcribe.
|
|
// TODO: use https://github.com/mccoyst/ogg?
|
|
func audioToWav(src, dst string) error {
|
|
command := []string{"ffmpeg", "-i", src, "-format", "s16le", "-ar", "16000", "-ac", "1", "-acodec", "pcm_s16le", dst}
|
|
out, err := runCommand(command)
|
|
if err != nil {
|
|
return fmt.Errorf("error: %w out: %s", err, out)
|
|
}
|
|
return nil
|
|
}
|
|
|
|
func Transcript(model whisper.Model, audiopath, language string, threads uint) (schema.Result, error) {
|
|
res := schema.Result{}
|
|
|
|
dir, err := os.MkdirTemp("", "whisper")
|
|
if err != nil {
|
|
return res, err
|
|
}
|
|
defer os.RemoveAll(dir)
|
|
|
|
convertedPath := filepath.Join(dir, "converted.wav")
|
|
|
|
if err := audioToWav(audiopath, convertedPath); err != nil {
|
|
return res, err
|
|
}
|
|
|
|
// Open samples
|
|
fh, err := os.Open(convertedPath)
|
|
if err != nil {
|
|
return res, err
|
|
}
|
|
defer fh.Close()
|
|
|
|
// Read samples
|
|
d := wav.NewDecoder(fh)
|
|
buf, err := d.FullPCMBuffer()
|
|
if err != nil {
|
|
return res, err
|
|
}
|
|
|
|
data := buf.AsFloat32Buffer().Data
|
|
|
|
// Process samples
|
|
context, err := model.NewContext()
|
|
if err != nil {
|
|
return res, err
|
|
|
|
}
|
|
|
|
context.SetThreads(threads)
|
|
|
|
if language != "" {
|
|
context.SetLanguage(language)
|
|
} else {
|
|
context.SetLanguage("auto")
|
|
}
|
|
|
|
if err := context.Process(data, nil, nil); err != nil {
|
|
return res, err
|
|
}
|
|
|
|
for {
|
|
s, err := context.NextSegment()
|
|
if err != nil {
|
|
break
|
|
}
|
|
|
|
var tokens []int
|
|
for _, t := range s.Tokens {
|
|
tokens = append(tokens, t.Id)
|
|
}
|
|
|
|
segment := schema.Segment{Id: s.Num, Text: s.Text, Start: s.Start, End: s.End, Tokens: tokens}
|
|
res.Segments = append(res.Segments, segment)
|
|
|
|
res.Text += s.Text
|
|
}
|
|
|
|
return res, nil
|
|
}
|