Add sound generation api (#9)

* Add missing attributes for VoiceResponseModel

* Updating module to point to forked repo

* Tidying up go.mod

* Adding missing voice settings

* Adding support for request stitching

* Adding support for request stitching

* Fix dup SharingOptions struct from merge

* Add Sound Generation API

* Fix: revert user-agent/package url to original
This commit is contained in:
Lachlan Laycock
2024-11-26 06:39:34 +01:00
committed by GitHub
parent c585531fae
commit db0a2e1760
8 changed files with 243 additions and 163 deletions

98
client/sound_gen.go Normal file
View File

@@ -0,0 +1,98 @@
package client
import (
"bytes"
"context"
"encoding/json"
"fmt"
"io"
"net/http"
"github.com/taigrr/elevenlabs/client/types"
)
// SoundGenerationWriter generates a sound effect from text and writes it to the provided writer.
// If durationSeconds is 0, it will be omitted from the request and the API will determine the optimal duration.
// If promptInfluence is 0, it will default to 0.3.
func (c Client) SoundGenerationWriter(ctx context.Context, w io.Writer, text string, durationSeconds, promptInfluence float64) error {
params := types.SoundGeneration{
Text: text,
PromptInfluence: 0.3, // default value
}
if promptInfluence != 0 {
params.PromptInfluence = promptInfluence
}
if durationSeconds != 0 {
params.DurationSeconds = durationSeconds
}
body, err := c.requestSoundGeneration(ctx, params)
if err != nil {
return err
}
defer body.Close()
_, err = io.Copy(w, body)
return err
}
// SoundGeneration generates a sound effect from text and returns the audio as bytes.
// If durationSeconds is 0, it will be omitted from the request and the API will determine the optimal duration.
// If promptInfluence is 0, it will default to 0.3.
func (c Client) SoundGeneration(ctx context.Context, text string, durationSeconds, promptInfluence float64) ([]byte, error) {
params := types.SoundGeneration{
Text: text,
PromptInfluence: 0.3, // default value
}
if promptInfluence != 0 {
params.PromptInfluence = promptInfluence
}
if durationSeconds != 0 {
params.DurationSeconds = durationSeconds
}
body, err := c.requestSoundGeneration(ctx, params)
if err != nil {
return nil, err
}
defer body.Close()
var b bytes.Buffer
_, err = io.Copy(&b, body)
if err != nil {
return nil, err
}
return b.Bytes(), nil
}
func (c Client) requestSoundGeneration(ctx context.Context, params types.SoundGeneration) (io.ReadCloser, error) {
url := c.endpoint + "/v1/sound-generation"
client := &http.Client{}
b, err := json.Marshal(params)
if err != nil {
return nil, err
}
req, err := http.NewRequestWithContext(ctx, http.MethodPost, url, bytes.NewBuffer(b))
if err != nil {
return nil, err
}
req.Header.Set("xi-api-key", c.apiKey)
req.Header.Set("User-Agent", "github.com/taigrr/elevenlabs")
req.Header.Set("accept", "audio/mpeg")
res, err := client.Do(req)
if err != nil {
return nil, err
}
if res.StatusCode != http.StatusOK {
res.Body.Close()
return nil, fmt.Errorf("unexpected status code: %d", res.StatusCode)
}
return res.Body, nil
}