Keyboard shortcuts

Press or to navigate between chapters

Press S or / to search in the book

Press ? to show this help

Press Esc to hide this help

CLI Reference

Status: Current Last updated: 2026-09-16 09:47 EDT

This page documents the current public batchalign3 CLI surface. For anything you are scripting against, confirm with batchalign3 <command> --help.

For detailed input/output patterns and mutation behavior per command, see Command I/O Parity.

Command shape

batchalign3 [GLOBAL OPTIONS] COMMAND [COMMAND OPTIONS] [PATHS...]

Global options go before the command name.

Global options

OptionMeaning
-v, -vv, -vvvIncrease verbosity
--workers NMaximum concurrent files per job (default: auto-tune; GPU commands default to 1). Auto-tune is (ram_total_mb / 16 GB).clamp(1, 8) for GPU-bound work.
--force-cpuDisable MPS/CUDA and force CPU-only models
--server URLRemote server URL. Env fallback: BATCHALIGN_SERVER
--override-media-cacheBypass the media analysis cache (audio tasks only; text NLP tasks are not cached at all)
--require-media-cacheRequire reusable evidence at cache-backed media stages; a miss fails instead of authorizing inference. Conflicts with both cache-override forms.
--override-media-cache-tasks TASKSBypass only named audio-evidence caches (comma-separated: forced_alignment, utr_asr, rev_asr_evidence, speaker_diarization_raw_evidence)
--debug-dir PATHDirectory for pipeline debug artifacts (CHAT/JSON fixtures for offline replay). Env fallback: BATCHALIGN_DEBUG_DIR
--memory-tier {small,medium,large,fleet}Override the auto-detected memory tier (forces worker bootstrap and memory budgets for that tier regardless of actual system RAM)
--timeout SECONDSOperator override for the audio-task transport timeout. For ASR, the default is DERIVED per request from the audio’s own duration (DecodeBudgetSeconds, crates/batchalign-types/src/worker_v2/requests.rs) plus a fixed margin, not a flat number; --timeout can only RAISE that derived ceiling, never lower it below what the request’s own decode budget needs. Forced alignment and speaker diarization still use a flat default (1800 = 30 min) unless overridden.
--tui / --no-tuiToggle full-screen TUI for server-backed jobs (DirectHost local runs stay on terminal progress bars)
--open-dashboard / --no-open-dashboardToggle browser auto-open for submitted server job pages (macOS only, interactive TTY only)
--engine-overrides JSONPer-engine PARAMETERS, as a {string:string} JSON object, e.g. {"qwen_model":"Qwen/Qwen3-ASR-0.6B-hf","qwen_device":"cpu"}. Forwarded to the worker as opaque knobs. The asr / fa / utr / translate keys additionally select an engine and beat the per-command flags; see “Engine selection” below. The payload is parsed once, while the command line is parsed, so an invalid one is rejected before anything runs. A bare engine name (--engine-overrides whisper) is reported as the wrong-flag mistake it is, naming --asr-engine and its siblings, rather than as malformed JSON.
--sequentialProcess files one at a time with a single worker. No memory gate, no server. Ideal for small jobs on laptops
--no-serverSkip auto-detection of a local server; force direct in-process execution

BA2 compatibility flags (--memlog, --mem-guard, --adaptive-workers, --pool, --shared-models, etc.) have been removed. If your scripts use them, remove them.

Engine selection

These are per-command flags, not global ones: they go AFTER the command name, like batchalign3 transcribe in/ --asr-engine paraformer.

OptionCommandsMeaning
--asr-engine NAMEtranscribe, benchmarkASR engine.
--fa-engine NAMEalignForced-alignment engine.
--utr-engine NAMEalignUtterance-timing-recovery engine. Only consulted with --utr.
--existing-wor-boundaries {preserve,rebuild-from-evidence}alignv0.4.0 option for prior %wor/main boundaries. Default preserve; rebuild mode is experimental and does not change raw FA cache identity.
--end-overlap-policy {clamp-all-adjacent,preserve-cross-speaker}alignSame-speaker (default, preserve-cross-speaker) or every adjacent pair (clamp-all-adjacent) end overlap resolved from measured word hulls. Cross-speaker overlap is left alone under the default, since it is ordinary conversation. No raw FA cache-key change.
--translate-engine NAMEtranslateTranslation engine.

Each flag’s --help lists every value it accepts, derived from the engine enum itself, so the advertised set and the accepted set are the same list. An unrecognized name is rejected while parsing, naming what you typed and suggesting the nearest valid value.

The older --asr-engine-custom / --fa-engine-custom / --utr-engine-custom flags still work and are hidden from help. They exist because each visible flag used to advertise only some of its engines, which left the rest reachable only through a second, differently-named flag: that is how the Cantonese engines stayed hidden from the people who needed them. Prefer the flags above.

The asr / fa / utr / translate keys of --engine-overrides also select an engine, and take precedence over the flags. Use the flags for ordinary work; the keys exist so one shared option can pin engines across a batch.

utr was missing from that set until 2026-08-06: it parsed, was forwarded to the Python worker as an opaque extra, and was ignored by everything, so a user who wrote it saw their choice silently dropped.

Sequential mode

--sequential gives you the simplest possible execution path, similar to batchalign2’s direct mode. One worker per task type, files processed one at a time, no concurrency infrastructure:

batchalign3 morphotag corpus/ -o output/ --sequential

What it does:

  • Forces --workers 1 and --no-server
  • Disables the memory gate (no cross-process coordination)
  • Keeps the worker alive for the entire run (no idle timeout kills)
  • Preserves the utterance cache (repeated runs benefit from cached results)

When to use it:

  • Processing a handful of files on a laptop
  • Debugging pipeline issues (predictable, single-threaded execution)
  • Environments where memory auto-tuning is unwanted

When NOT to use it:

  • Large corpus runs (50+ files), the default parallel mode is 3-5× faster
  • Fleet machines with warm workers, use the server instead

--sequential is incompatible with --server (mutually exclusive).

Dashboard browser auto-open

On macOS, when you run a processing command interactively (e.g., batchalign3 transcribe corpus/ output/), the CLI automatically opens the job’s dashboard page in your default browser. This lets you monitor progress in real time.

Direct local execution does not submit an HTTP job, so there is no dashboard page to open. In that mode, --open-dashboard is a no-op and the CLI shows local terminal progress inline instead.

The dashboard auto-open is only triggered when:

  • Running on macOS (no-op on Linux/Windows)
  • stderr is connected to an interactive terminal (TTY)
  • --no-open-dashboard was not passed
  • The BATCHALIGN_NO_BROWSER environment variable is not set

It will not fire in non-interactive contexts: cron jobs, CI pipelines, SSH sessions without a display, piped output, or scripts. To suppress it explicitly in interactive sessions, pass --no-open-dashboard.

Common path-processing options

The core processing commands documented below all accept:

OptionMeaning
PATHS...Input files or directories
-o, --output DIROutput directory
--file-list FILERead input paths from a text file (see below)
--in-placeModify inputs in place

When exactly two positional paths are provided, the CLI still accepts the legacy input/output directory form. For new scripts, prefer -o/--output.

--file-list format

--file-list FILE reads input paths from a plain-text UTF-8 file, one path per line:

  • Blank lines and lines beginning with # are ignored; whitespace around each path is trimmed.
  • A relative path resolves against the directory containing the list file, not the directory you run the command from. Absolute paths are used as written.
  • A path naming a directory is expanded exactly like a positional directory argument: matching files are discovered recursively beneath it.
  • The input set is de-duplicated, keeping the first occurrence. A file listed twice, spelled two ways (a.cha, ./a.cha, an absolute path), or reached both directly and through a listed directory is processed once.
  • Every entry must exist when the command runs. A missing entry is a usage error (exit code 2) naming the list file and line, for example lists/rerun.txt:3: input path does not exist: lists/corpus/missing.cha.
# lists/rerun.txt: entries are relative to lists/
corpus/session-01.cha
corpus/session-02.cha

# a whole directory, and an absolute path
corpus/follow-up/
/data/project/extra/session-09.cha
# Run align on every listed input (in place, against a remote server)
batchalign3 --server http://your-server:8001 align --file-list lists/rerun.txt

To process a large list in smaller batches, split it into chunk files in the same directory as the original list (for example, run split -l 10 rerun.txt batch- inside that directory) so relative entries keep resolving against the same place, then run batchalign3 align --file-list <chunk> on each chunk sequentially.

--file-list cannot be combined with positional PATHS arguments; the CLI rejects the combination. Without -o/--output, every listed input is processed in place (output overwrites input); with -o DIR, results are written under DIR exactly as for positional inputs.

For batched text-NLP commands (morphotag, utseg, translate, coref), large --file-list runs may not show file-by-file on-disk rewrites while the invocation is still running. The command can batch/stage work internally and then commit the in-place writes when the current invocation finishes. If you need visible write-through during a long rerun, split the list into smaller chunks and run those chunks sequentially.

Processing commands

Each processing command has a dedicated page with full options, a pipeline diagram, examples, and gotchas. Click the command name for complete documentation.

CHAT-mutation commands (input .cha → output .cha)

CommandWhat it does
alignAdd word-level and utterance-level timestamps via forced alignment
morphotagAdd %mor POS/lemma and %gra dependency tiers
utsegRe-segment utterance boundaries using Stanza constituency parsing
translateAdd %xtra English translation tiers
corefAdd sparse %xcoref coreference annotation tiers (English only)
compareCompare against gold .cha references; write %xsrep/%xsmor + .compare.csv

Audio-input commands (input audio → new files)

CommandWhat it does
transcribeCreate .cha transcripts from audio via ASR
benchmarkTranscribe and evaluate WER against gold .cha references
opensmileExtract acoustic features → .opensmile.csv (positional I/O)
avqiCalculate Acoustic Voice Quality Index from paired .cs/.sv audio (positional I/O)

Operational commands

setup

Initialize ~/.batchalign.ini:

batchalign3 setup
batchalign3 setup --non-interactive --engine whisper
batchalign3 setup --non-interactive --engine rev --rev-key <KEY>

Options:

OptionMeaning
--engine {rev,whisper}Persist default ASR engine
--rev-key KEYRev.AI key for non-interactive setup
--non-interactiveDisable prompts

logs

batchalign3 logs
batchalign3 logs --last
batchalign3 logs --export
batchalign3 logs --clear

Key options:

OptionMeaning
--lastShow the most recent run log
--rawRaw JSONL output with --last
--exportZip recent logs
--clearDelete log files
--followTail the newest log file
-n, --count NNumber of recent runs to list

serve

batchalign3 serve start --foreground
batchalign3 serve status
batchalign3 serve stop

serve start key options:

OptionMeaning
--port PORTListen port
--host HOSTBind address
--config PATHAlternate server.yaml path
--python PATHWorker Python executable
--foregroundDo not daemonize
--test-echoStart test-echo workers

jobs

batchalign3 jobs --server http://myserver:8000
batchalign3 jobs --server http://myserver:8000 <JOB_ID>
batchalign3 jobs <JOB_ID>
batchalign3 jobs --json <JOB_ID>
batchalign3 jobs cancellations <JOB_ID>

With --server, lists or inspects remote jobs. Without --server, inspects the local job artifact directory for post-failure debugging. Pass --json for machine-readable output.

The cancellations subcommand prints the cancellation audit history for a single job, every cancel attempt is recorded with source (tui / api / dashboard / staging / signal), host, pid, reason, and in_flight_filename. Use this when a user reports “I didn’t cancel that job.”

cache

batchalign3 cache stats
batchalign3 cache clear --yes
batchalign3 cache clear --all --yes

BATCHALIGN_ANALYSIS_CACHE_DIR and BATCHALIGN_MEDIA_CACHE_DIR relocate the underlying caches for isolated runs. BA2-compatible flag forms cache --stats and cache --clear are still accepted.

openapi

batchalign3 openapi -o openapi.json
batchalign3 openapi --check --output openapi.json

--check exits non-zero when the target file does not match the generated schema.

models

Two subcommands:

SubcommandPurpose
models prepExtract training text from CHAT files (Rust-native, no CLAN needed)
models trainForward to the Python training runtime (python -m batchalign.models.training.run)

See Models Training Runtime ADR.

ipc-schema

batchalign3 ipc-schema -o schemas/
batchalign3 ipc-schema --check --output schemas/

Emits JSON Schema for Rust→Python IPC types. Without -o, schemas are written to stdout as a single JSON object. With --check, exits non-zero on schema drift against the target directory.

bench

batchalign3 bench <COMMAND> <IN_DIR> <OUT_DIR> [--runs N]
batchalign3 bench align corpus/ out/ --runs 3 --dataset eng-childes-v1
batchalign3 bench align corpus/ out/ --use-cache

Benchmark command execution time across repeated runs. <COMMAND> is one of: align, transcribe, transcribe_s (with diarization), morphotag, translate, utseg, benchmark, opensmile, coref, compare. Distinct from the benchmark top-level command, which measures ASR word accuracy.

OptionMeaning
--runs NNumber of repeat runs (default: 1)
--dataset LABELDataset label included in structured output (useful for cross-run comparison)
--use-cacheUse the analysis cache for benchmark runs (default: bypass cache so each run hits cold paths)

doctor

batchalign3 doctor
batchalign3 doctor --lang yue --format json
batchalign3 doctor --explain memory_gate_mb
batchalign3 doctor --warnings-as-errors

Pre-flight diagnostic that spawns a test worker, sends known inputs through the morphosyntax pipeline, and validates the output structure. Catches machine-specific issues (stale models, missing processors, MWT quirks) before they become production failures.

OptionMeaning
--lang LANGLanguage to test (default: eng)
--format {human,json}Output format (default: human)
--python PATHCustom Python path (overrides BATCHALIGN_PYTHON)
--explain KNOBTrace why one resolved knob has its current value (gpu_thread_pool_size, force_cpu, max_total_workers, max_concurrent_jobs, max_workers_per_key, memory_gate_mb). Prints resolved value, source (operator override vs. host-facts recommendation), the rule that produced the recommendation, and the relevant detected facts. Implies --check.
--warnings-as-errorsTreat host-facts validation warnings as fatal: exit non-zero when any warning fires, not only on error. Intended for CI gates that want zero-warning deployments. Has no effect outside --check / --explain.

replay

batchalign3 replay <DUMP_FILE>
batchalign3 replay --lang yue path/to/failed_ipc_*.json

Replay a captured failed IPC request against a fresh worker. Takes a dump file from ~/.batchalign3/debug/ and sends the exact request to a new worker, reporting the response. Useful for reproducing field failures locally.

eval

batchalign3 eval l2-morphotag <ARGS>
batchalign3 eval utr-alignment --chat <CHAT> --tokens <JSON> --output <JSON>
batchalign3 eval utseg-replay post-chat --input-chat <CHAT> --evidence <JSON> --output-chat <CHAT>
batchalign3 eval utseg-replay pre-asr --asr-response <JSON> --evidence <JSON> --output-chat <CHAT> [--media-name <NAME>] [--wor]

Evaluation subcommands. Currently:

SubcommandPurpose
eval l2-morphotagL2 morphotag evaluation: pair @s words with %mor / %gra items via typed AST walk (supersedes scripts/l2-eval/analyze.py)
eval utr-alignmentOffline global UTR word-to-token replay with fingerprinted typed evidence and no inference or CHAT mutation
eval utseg-replayReapply retained utterance-boundary evidence and report whether it still reproduces the document the run wrote; exits 1 on a difference, 2 on a refused input

version

batchalign3 version

Prints version and build information.

Exit codes

batchalign3 uses stable non-zero exit code categories:

CodeMeaning
2Usage/input error
3Configuration error
4Network/connectivity error
5Server/job lifecycle error
6Local runtime error

Exit code 1 is reserved for unexpected failures outside the typed categories.

A server that reports another build than the CLI’s, or no build, is refused before anything is submitted, with exit code 5; see Server Mode: build identity check.


This page last changed: 2026-09-16 (commit 197c81e6). The whole book last changed: 2026-09-16 (commit 34d249d8).