feat: streaming audio endpoint with MinIO write-through cache
Some checks failed
CI / Backend (push) Failing after 11s
Release / Check ui (push) Failing after 55s
Release / Test backend (push) Successful in 56s
Release / Docker / ui (push) Has been skipped
CI / UI (push) Failing after 1m6s
Release / Docker / caddy (push) Successful in 41s
CI / Backend (pull_request) Successful in 58s
CI / UI (pull_request) Successful in 55s
Release / Docker / runner (push) Failing after 55s
Release / Docker / backend (push) Failing after 1m23s
Release / Gitea Release (push) Has been skipped

Add GET /api/audio-stream/{slug}/{n}?voice= that streams MP3 audio to the
client as TTS generates it, while simultaneously uploading to MinIO. On
subsequent requests the endpoint redirects to the presigned MinIO URL,
skipping generation entirely.

- PocketTTS: StreamAudioMP3 pipes live WAV response body through ffmpeg
  (streaming transcode — no full-buffer wait)
- Kokoro: StreamAudioMP3 uses stream:true mode, returning MP3 frames
  directly without the two-step download-link flow
- AudioStore: PutAudioStream added for multipart MinIO upload from reader
- WriteTimeout bumped 60s → 15min to accommodate full-chapter streams
- X-Accel-Buffering: no header disables Caddy/nginx response buffering
This commit is contained in:
Admin
2026-03-30 22:02:36 +05:00
parent b0a4cb8b3d
commit 4e7f8c6266
11 changed files with 330 additions and 28 deletions

View File

@@ -155,6 +155,14 @@ func (m *minioClient) putObject(ctx context.Context, bucket, key, contentType st
return err
}
// putObjectStream uploads from r with known size (or -1 for multipart).
func (m *minioClient) putObjectStream(ctx context.Context, bucket, key, contentType string, r io.Reader, size int64) error {
_, err := m.client.PutObject(ctx, bucket, key, r, size,
minio.PutObjectOptions{ContentType: contentType},
)
return err
}
func (m *minioClient) getObject(ctx context.Context, bucket, key string) ([]byte, error) {
obj, err := m.client.GetObject(ctx, bucket, key, minio.GetObjectOptions{})
if err != nil {

View File

@@ -4,6 +4,7 @@ import (
"context"
"encoding/json"
"fmt"
"io"
"log/slog"
"strings"
"time"
@@ -383,6 +384,10 @@ func (s *Store) PutAudio(ctx context.Context, key string, data []byte) error {
return s.mc.putObject(ctx, s.mc.bucketAudio, key, "audio/mpeg", data)
}
func (s *Store) PutAudioStream(ctx context.Context, key string, r io.Reader, size int64, contentType string) error {
return s.mc.putObjectStream(ctx, s.mc.bucketAudio, key, contentType, r, size)
}
// ── PresignStore ──────────────────────────────────────────────────────────────
func (s *Store) PresignChapter(ctx context.Context, slug string, n int, expires time.Duration) (string, error) {