dearwolf/LocalAI
1
0
Fork 0
mirror of https://github.com/mudler/LocalAI.git synced 2025-05-30 23:44:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 34

feat(tts): Implement naive response_format for tts endpoint (#4035)

Browse source 
Signed-off-by: n-Arno <arnaud.alcabas@gmail.com>
This commit is contained in:
Arnaud A 2024-11-02 20:13:35 +01:00 • committed by GitHub
parent 57908df956
commit 65c3df392c
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
6 changed files with 61 additions and 9 deletions
Download patch file Download diff file Expand all files Collapse all files

24
core/http/endpoints/localai/tts.go
View file

@ -9,16 +9,19 @@ import (
"github.com/gofiber/fiber/v2"
"github.com/mudler/LocalAI/core/schema"
"github.com/rs/zerolog/log"
"github.com/mudler/LocalAI/pkg/utils"
)
// TTSEndpoint is the OpenAI Speech API endpoint https://platform.openai.com/docs/api-reference/audio/createSpeech
// @Summary Generates audio from the input text.
// @Accept json
// @Produce audio/x-wav
// @Param request body schema.TTSRequest true "query params"
// @Success 200 {string} binary "generated audio/wav file"
// @Router /v1/audio/speech [post]
// @Router /tts [post]
//
// @Summary Generates audio from the input text.
// @Accept json
// @Produce audio/x-wav
// @Param request body schema.TTSRequest true "query params"
// @Success 200 {string} binary "generated audio/wav file"
// @Router /v1/audio/speech [post]
// @Router /tts [post]
func TTSEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig) func(c *fiber.Ctx) error {
return func(c *fiber.Ctx) error {
@ -67,6 +70,13 @@ func TTSEndpoint(cl *config.BackendConfigLoader, ml *model.ModelLoader, appConfi
if err != nil {
return err
}
// Convert generated file to target format
filePath, err = utils.AudioConvert(filePath, input.Format)
if err != nil {
return err
}
return c.Download(filePath)
}
}

1
core/schema/localai.go
View file

@ -32,6 +32,7 @@ type TTSRequest struct {
Voice string `json:"voice" yaml:"voice"` // voice audio file or speaker id
Backend string `json:"backend" yaml:"backend"`
Language string `json:"language,omitempty" yaml:"language,omitempty"` // (optional) language to use with TTS model
Format string `json:"response_format,omitempty" yaml:"response_format,omitempty"` // (optional) output format
}
type StoresSet struct {