Files
silo-server/internal/playback/audio_extract.go
T
39ba284c9d feat(ai): shared AI core — metadata translation, Whisper ASR, per-profile language, on-view translation (#127)
* docs: design + plan for shared AI core, metadata translation, Whisper ASR

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* refactor(ai): shared LLM client, segment translator, and job runner packages

internal/ai/llm: OpenAI-compatible chat client moved out of subtitles/ai,
plus /v1/audio/transcriptions (verbose_json) for the ASR work; one shared
retry/backoff loop for both. internal/ai/translate: the batched indexed-JSON
translation protocol generalized to text segments. internal/ai/jobrunner:
dispatch/heartbeat/reaper/cancel lifecycle extracted behind a minimal store
interface, with a semaphore shareable across job services.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* refactor(subtitles): consume shared AI core

LLMTranslator becomes a thin cue<->segment adapter over aitranslate; the
service delegates dispatch/heartbeat/reaper/cancel to jobrunner; the local
OpenAI client is gone in favor of internal/ai/llm. Behavior (prompts, wire
protocol, job rows, recovery semantics) is unchanged. NewService now takes
the dispatch semaphore so all AI job services can share one bound.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* feat(config): shared ai.* settings, metadata translation job table, localization provenance columns

ai.* connection keys (chat + optional separate ASR endpoint) load with a
fallback to the legacy subtitle_ai.* rows — those are never renamed in SQL
because encrypted values are GCM-bound to their setting key. New toggles:
subtitle_ai.transcribe_enabled, metadata_ai.enabled. Migration adds
metadata_translation_jobs, per-field provenance (provider|ai|manual) on the
localization tables, and media_folders.auto_translate_metadata.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* feat(catalog): localization field provenance with provider/ai/manual precedence

Provider upserts keep manual values and never blank a field with an empty
incoming value; new UpsertAITranslation/UpsertAIOverview methods write AI
fields only over empty or ai-sourced values (force adds provider, never
manual) — all enforced in single-statement SQL. Serving now merges only
non-empty localized fields onto the base item, since localization rows are
legitimately partial (AI rows carry no titles/artwork).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* feat(metadata): AI translation service, refresh auto-fallback, and admin API

internal/metadata/translation: job service over the shared AI core that
expands an item to its season/episode overviews, skips already-localized
fields (zero model calls on repeat runs), batches paragraphs through the
generic translator, and persists per batch with provenance-aware upserts.
MetadataService gains an AutoTranslator seam invoked after each refresh for
libraries with auto_translate_metadata. Admin endpoints under the metadata
curation guard: enqueue, list (poll), cancel; plus a status probe.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* feat(subtitles): Whisper ASR transcribe and transcribe_translate jobs

New WhisperTranscriber: one ffmpeg pass extracts the audio track to 10-min
16kHz mono WAV chunks (temp dir cleaned on every exit path), each chunk goes
to the OpenAI-compatible /v1/audio/transcriptions endpoint (verbose_json,
per-request timeout sized to 3x chunk duration), segment timestamps are
offset and built into wrapped cues. Chunks process playhead-first and stream
live to the requesting session. The transcript is stored as an ordinary
downloaded subtitle (provider 'transcribed'); transcribe_translate chains
the existing translator and stores the translated track as the job result.
Enqueue accepts an optional kind; status reports transcribe_enabled.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* feat(web): AI services settings, metadata translate action, library auto-translate, generate-from-audio

New AI Services admin page hosts the shared endpoint config (reads fall back
to legacy subtitle_ai.* values, writes target ai.*) and the three feature
toggles; the AI card moves out of Subtitles settings. The metadata editor
gains a Translate-with-AI panel with job polling and force/re-translate. The
library form gains the auto-translate toggle (threaded through the libraries
API). The player translate modal gains a From-audio mode that lists audio
tracks and submits transcribe / transcribe_translate jobs.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* style: gofmt import grouping in router and translation tests

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* feat(catalog): per-profile metadata language and viewer-triggered description translation

user_profiles.preferred_metadata_language threads through the access scope
into catalog serving: presentation language now resolves explicit param ->
profile preference -> library metadata language (native API and jellycompat).
ItemDetail gains pending_translation_language when the viewer's language is
missing a localized overview. New metadata_ai.on_view setting (off|button|
auto) gates POST /items/{id}/translate-description: any profile with item
access may request its language, with in-flight dedup and a 15-minute
failure cooldown so page views never hammer a broken endpoint.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* feat(web): on-view description translation with per-profile metadata language

Profile playback settings gain a Metadata language picker (library default
inherit). Detail pages: when the server reports pending_translation_language
and metadata_ai.on_view is 'auto', the description translates on view with a
pulse animation until the refetched detail comes back localized (45s
timeout); in 'button' mode a small Translate chip triggers the same flow.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* fix(web): expose metadata_ai.on_view in AI Services settings

The on-view translation mode had no UI control, so it could only ever be
'off' — viewers got neither the auto translation nor the fallback button.
Adds the off/button/auto selector to the Features card, and the config
loader now warns and falls back to 'off' on a bad row instead of refusing
to start.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* fix(ai): clear configuration hint when the transcription endpoint is chat-only

A blank Transcription base URL falls back to the chat endpoint; chat-only
gateways reject the multipart upload with an opaque 400 that reads like a
pipeline bug. 400/404/405 transcription failures now carry a hint to set a
Whisper-compatible endpoint in AI Services.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* fix(subtitles): wrap ASR cue text by rune count, not bytes

Arabic/Cyrillic/Greek text is 2+ bytes per character in UTF-8, so byte-based
wrapping broke lines at roughly half the intended visual width.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* docs(web): steer transcription base URL hint away from chat-only gateways

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* feat(ai): block chat-only gateways for transcription, add endpoint presets

llm.IsChatOnlyGateway (OpenRouter et al — no timestamped transcription API)
is enforced in three layers: the settings API rejects ai.asr_base_url values
pointing at one, the router disables ASR with a warning when the blank-URL
fallback would land on one, and llm.Transcribe refuses outright. The AI
Services page gains one-click transcription presets (Groq turbo/accurate,
OpenAI, self-hosted speaches) plus the mirrored client-side check, and the
settings API now also validates metadata_ai.on_view.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* fix(subtitles): tighten ASR subtitle sync

Three systematic timing-error sources addressed: cue offsets now use the
segment muxer's exact per-chunk start times (segment_list CSV) instead of
assuming index*chunk_seconds; the audio stream's start delay relative to the
container timeline (common in TS remuxes) is probed via ffprobe and added to
every cue; and the chunk length is now operator-tunable via
subtitle_ai.asr_chunk_seconds (60-600s, default 600) since shorter chunks
bound Whisper's within-chunk timestamp drift. Playhead-first ordering now
pivots on real chunk starts, and a beyond-end playhead starts at the final
chunk instead of restarting from zero.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* fix(ai): tolerate base URLs that already include the /v1 segment

Providers like DeepInfra expose their OpenAI-compatible API under a base
that contains the version segment (api.deepinfra.com/v1/openai); always
appending /v1/... mangled those. endpointURL now appends bare paths when
the base already carries /v1.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* feat(web): prefer self-hosted transcription in presets and hints

Preset order becomes self-hosted (recommended) -> Groq turbo -> Groq
large-v3 -> OpenAI, and the settings hint plus the job-error hint lead with
the self-hosted option. The self-hosted preset now fills the turbo CT2 model
to match the recommended speaches setup.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* fix(subtitles): request VAD and word timestamps for ASR cue accuracy

Without vad_filter, faster-whisper servers report wall-to-wall segment
times: cues linger on screen through silence (verified up to 91s) and
paragraph-length segments become single 400+ char cues. Request
vad_filter=true (skipped for hosted providers that reject non-OpenAI
fields and run VAD server-side) plus timestamp_granularities word+segment,
and rebuild cues from word timings: split at speech pauses, sentence ends,
text capacity, and a 7s max duration; cap word-less segments instead of
trusting their reported end; stretch sub-second cues to a readable minimum.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
2026-06-10 14:58:54 -04:00

181 lines
5.2 KiB
Go

package playback
import (
"bytes"
"context"
"encoding/csv"
"encoding/json"
"fmt"
"math"
"os"
"os/exec"
"path/filepath"
"sort"
"strconv"
"strings"
)
// AudioChunk is one extracted piece of an audio track. Start is the chunk's
// exact start within the extracted audio (seconds) as reported by ffmpeg's
// segment muxer — the muxer cuts at packet boundaries near, not exactly at,
// the requested length, so assuming index*chunkSeconds would accumulate
// subtitle timing error.
type AudioChunk struct {
Path string
Start float64
}
// ExtractAudioChunks extracts one audio track from a media file into
// fixed-length 16 kHz mono WAV chunks under dir — the input format Whisper
// endpoints want, sized to stay under typical upload limits (a 10-minute
// chunk is ~19 MB). One ffmpeg pass segments the whole track; the returned
// chunks are in chronological order with exact start offsets. The caller owns
// dir and its cleanup.
func ExtractAudioChunks(ctx context.Context, filePath string, audioTrackIndex int, dir, ffmpegPath string, chunkSeconds int) ([]AudioChunk, error) {
if chunkSeconds <= 0 {
chunkSeconds = 600
}
if audioTrackIndex < 0 {
audioTrackIndex = 0
}
listPath := filepath.Join(dir, "segments.csv")
args := []string{
"-i", filePath,
"-vn", "-sn", "-dn",
"-map", fmt.Sprintf("0:a:%d", audioTrackIndex),
"-ac", "1",
"-ar", "16000",
"-c:a", "pcm_s16le",
"-f", "segment",
"-segment_time", strconv.Itoa(chunkSeconds),
"-segment_list", listPath,
"-segment_list_type", "csv",
"-y", filepath.Join(dir, "chunk%05d.wav"),
}
ffmpeg := "ffmpeg"
if ffmpegPath != "" {
ffmpeg = ffmpegPath
}
cmd := exec.CommandContext(ctx, ffmpeg, args...)
var stderr bytes.Buffer
cmd.Stderr = &stderr
if err := cmd.Run(); err != nil {
return nil, fmt.Errorf("ffmpeg audio extraction failed: %w (stderr: %s)",
err, truncateStderr(stderr.String()))
}
starts := parseSegmentList(listPath)
entries, err := os.ReadDir(dir)
if err != nil {
return nil, fmt.Errorf("read audio chunk dir: %w", err)
}
var names []string
for _, e := range entries {
if !e.IsDir() && strings.HasPrefix(e.Name(), "chunk") && strings.HasSuffix(e.Name(), ".wav") {
names = append(names, e.Name())
}
}
sort.Strings(names)
if len(names) == 0 {
return nil, fmt.Errorf("ffmpeg produced no audio chunks for track %d", audioTrackIndex)
}
chunks := make([]AudioChunk, len(names))
for i, name := range names {
start, ok := starts[name]
if !ok {
// Segment list missing/unparsable: fall back to the nominal grid.
start = float64(i * chunkSeconds)
}
chunks[i] = AudioChunk{Path: filepath.Join(dir, name), Start: start}
}
return chunks, nil
}
// parseSegmentList reads ffmpeg's CSV segment list (filename,start,end per
// line) into a filename → start map. Best effort: a missing or malformed list
// yields an empty map and callers fall back to nominal chunk starts.
func parseSegmentList(listPath string) map[string]float64 {
starts := map[string]float64{}
f, err := os.Open(listPath)
if err != nil {
return starts
}
defer f.Close()
rows, err := csv.NewReader(f).ReadAll()
if err != nil {
return starts
}
for _, row := range rows {
if len(row) < 2 {
continue
}
start, err := strconv.ParseFloat(strings.TrimSpace(row[1]), 64)
if err != nil || start < 0 {
continue
}
starts[filepath.Base(strings.TrimSpace(row[0]))] = start
}
return starts
}
// maxPlausibleAudioDelay caps the probed audio/container start delta; beyond
// this it is far more likely a probing artifact than a real stream delay.
const maxPlausibleAudioDelay = 30.0
// ProbeAudioStartOffset returns the audio stream's start time relative to the
// container start (seconds). Whisper timestamps are relative to the first
// audio sample, while the playback timeline starts at the container start —
// in containers with delayed audio (TS remuxes especially) the difference is
// a constant subtitle sync error unless corrected. Best effort: any probe
// failure returns 0.
func ProbeAudioStartOffset(ctx context.Context, filePath string, audioTrackIndex int, ffmpegPath string) float64 {
if audioTrackIndex < 0 {
audioTrackIndex = 0
}
ffprobe := "ffprobe"
if ffmpegPath != "" {
// ffprobe ships next to ffmpeg in every distribution Silo supports.
ffprobe = filepath.Join(filepath.Dir(ffmpegPath), "ffprobe")
}
cmd := exec.CommandContext(ctx, ffprobe,
"-v", "error",
"-print_format", "json",
"-show_entries", "format=start_time",
"-select_streams", fmt.Sprintf("a:%d", audioTrackIndex),
"-show_entries", "stream=start_time",
filePath,
)
out, err := cmd.Output()
if err != nil {
return 0
}
var probe struct {
Format struct {
StartTime string `json:"start_time"`
} `json:"format"`
Streams []struct {
StartTime string `json:"start_time"`
} `json:"streams"`
}
if err := json.Unmarshal(out, &probe); err != nil || len(probe.Streams) == 0 {
return 0
}
formatStart, err1 := strconv.ParseFloat(probe.Format.StartTime, 64)
streamStart, err2 := strconv.ParseFloat(probe.Streams[0].StartTime, 64)
if err1 != nil || err2 != nil {
return 0
}
delta := streamStart - formatStart
if math.IsNaN(delta) || math.Abs(delta) > maxPlausibleAudioDelay {
return 0
}
return delta
}