Environment variables

August 20, 2026 · View on GitHub

CrispASR exposes a large number of environment variables — tuning knobs, device placement overrides, benchmark timers, and debug/dump switches. This page is the single reference for all of them: the naming convention, the cross-cutting variables, the per-backend variables, and the ones CrispASR deliberately does not own.

This document was introduced for issue #265 ("consistency and documentation"). The same change standardized every backend-owned variable onto one prefix and unified the reference-voice cache location (see below).

Naming convention

Every CrispASR-owned variable is named:

CRISPASR_<BACKEND>_<FEATURE>

for example CRISPASR_OMNIVOICE_CODEC_GPU, CRISPASR_IRODORI_CODEC_GPU, CRISPASR_PARAKEET_BENCH. Process-global variables that are not tied to one backend drop the <BACKEND> segment (CRISPASR_MODELS_DIR, CRISPASR_KV_ON_CPU).

Legacy aliases and deprecation warnings

Historically many backends used a bare prefix (OMNIVOICE_CODEC_GPU, QWEN3_TTS_BENCH, CHATTERBOX_DEBUG, …). Those bare names still work as legacy aliases so existing scripts, notebooks, and Kaggle A/B kernels keep running. The alias is derived automatically by stripping the CRISPASR_ prefix, so for any CRISPASR_FOO_BAR documented here the bare FOO_BAR is also honored.

The first time a run reads a value from a legacy name, CrispASR prints a one-time notice to stderr:

[crispasr] warning: environment variable 'OMNIVOICE_CODEC_GPU' is deprecated and
will be removed in a future release; use 'CRISPASR_OMNIVOICE_CODEC_GPU' instead
(the old name still works for now). ...

Migrate to the canonical CRISPASR_-prefixed name at your convenience. To silence the warnings (e.g. for a pipeline that scrapes stderr and cannot migrate yet), set:

CRISPASR_SUPPRESS_ENV_DEPRECATION=1

The lookup + aliasing is implemented once in src/core/crispasr_env.h (crispasr_env::get / truthy / present); new backends should read env vars through that helper with a canonical CRISPASR_-prefixed name.

Value/truthiness conventions

Most variables are read with one of two conventions — check the surrounding suffix legend, but as a rule:

  • Flag (opt-in) — set to 1 (or any non-empty, non-0 value) to enable. Some legacy flags treat being set at all as on, so prefer =1 / unset it entirely rather than =0 when a flag's exact semantics matter.
  • Path / value — the variable's value is a filesystem path, integer, or float used directly.

Common suffix legend

The same feature suffixes recur across nearly every backend. Rather than repeat them per backend below, they mean:

SuffixMeaning
_BENCHPrint per-stage wall-clock timings for that engine.
_DEBUG / _DIAGVerbose debug logging.
_DUMP_DIR / _DUMP / _DUMP_*Write intermediate tensors / stages to disk (diagnostics, diff-harness).
_FORCE_SCALARDisable SIMD kernels (numeric-parity debugging).
_GPU / _USE_GPU / _FORCE_METALForce that stage/codec onto the GPU.
_CPU / _CPU_ONLY / _FORCE_CPUForce that stage/codec onto the CPU.
_NO_FA / _FLASH_ATTNToggle flash-attention.
_FASTCONV (_DEBUG)Enable the baked-F32 / matmul conv fast path for a codec/vocoder.
_CFG_INTERVAL (_DEBUG)Interval-CFG cadence for a flow-matching decoder.
_SEEDRNG seed for reproducible generation.
_*_FILE / _*_PATH / _*_GGUFOverride an asset/model/fixture path (mostly dev/test).

Global / cross-cutting variables

These are not tied to a single backend.

Paths, cache, and models

VariablePurpose
CRISPASR_MODELS_DIRDirectory searched for GGUF models (also the auto-download target root).
CRISPASR_CACHE_DIRBase cache directory for auto-downloaded models/assets (default ~/.cache/crispasr).
CRISPASR_SCRATCH_DIRScratch directory for temporary run artifacts.
CRISPASR_DUMP_DIRGlobal tensor-dump directory (diagnostics).
CRISPASR_GGUF_MMAP / CRISPASR_GGUF_PRELOADControl GGUF mmap vs. preload-into-RAM loading.
CRISPASR_MLOCKmlock model weights into RAM.
CRISPASR_IGNORE_CPU_ISAContinue past the startup build-vs-host CPU instruction-set check (#380) instead of exiting; the process will SIGILL at the first compute op the CPU can't run.

GPU / device placement

VariablePurpose
CRISPASR_N_GPU_LAYERSNumber of transformer layers to offload to the GPU.
CRISPASR_ARG_DEVICEDefault device selection for the CLI.
CRISPASR_KV_ON_CPUKeep the KV cache on the CPU.
CRISPASR_KV_QUANT / _KV_QUANT_K / _KV_QUANT_V / _KV_READ_F32KV-cache quantization / read format.

Device selection across compiled backends also honors the standard ggml / CUDA variables CUDA_VISIBLE_DEVICES and GGML_VK_VISIBLE_DEVICES — see "Variables CrispASR does not own" below.

Session / long-audio chunking

VariablePurpose
CRISPASR_SESSION_AUTOCHUNKEnable auto-chunking in the session API for long audio.
CRISPASR_SESSION_CHUNK_SECONDSChunk length (seconds) for session auto-chunking.
CRISPASR_SESSION_PERBACKEND_CHUNKUse per-backend chunk-window tuning instead of a flat window.
CRISPASR_SESSION_UNIFIED_DISPATCHRoute surfaces through the unified library dispatch path.

Post-decode hygiene (PLAN.md §W2–W7)

All OFF unless set. Each of these can delete or alter user-visible text, so none of them switches on by surprise; a wrong deletion is worse than a surviving artifact. Applied on both the CLI and the session C-ABI.

VariablePurpose
CRISPASR_SEG_MAX_CHARSTruncate any segment longer than N code points, backing up to the last 。．.！!？?、, but never below 75% of the cap. A line past the cap is almost always a repetition hallucination that survived the n-gram collapse.
CRISPASR_SEG_DROP_NONVERBALDrop segments that are entirely a non-verbal marker — [Music], (applause), （喘ぎ声）, ♪. Running speech merely containing such a word is never dropped.
CRISPASR_SEG_LOGPROB_THOLDDrop segments whose average log-probability is below this. Post-hoc, on top of the decoder's own fallback gate.
CRISPASR_SEG_LOGPROB_MARGINLoosen that threshold by this much for segments ≤1.6 s — a short segment's mean logprob is noisier, so it gets more room, not less.
CRISPASR_SEG_MERGE_REPEATSCollapse runs of near-identical adjacent segments into one spanning the whole run. Catches a phrase repeating across segment boundaries, which per-segment loop fixes cannot see.
CRISPASR_SEG_MERGE_SIMILARITYSimilarity bar for the above (default 0.90; LCS over code points).
CRISPASR_SEG_MERGE_GAP_CSNever merge across a gap wider than this many centiseconds (default 200). Two identical lines a minute apart are two real utterances.
CRISPASR_SEG_MERGE_MIN_RUNMinimum consecutive similar segments before merging (default 3), so an ordinary repeated "yes." pair survives.

Alignment and VAD sanity checks

VariablePurpose
CRISPASR_ALIGN_SENTINEL0 disables the forced-alignment collapse check. On by default, detect + warn only. Catches ctc_forced_align() returning words at t0 == t1 == 0 — its two silent-zero paths (characters absent from the CTC vocab, or a word the Viterbi path never visited) produce garbage timestamps inside a successful return.
CRISPASR_ALIGN_SENTINEL_REDISTRIBUTE1 opts into repair: respace the words across the clip in proportion to character count. Off by default — a wrong auto-repair would be just as invisible as the collapse.
CRISPASR_VAD_FAILOVER0 disables the VAD sanity check. On by default: if a clip over 120 s comes back with under 1% speech coverage (or a couple of segments covering under 10% of a very long clip), the VAD is wrong and the run falls back to fixed full-clip chunks rather than losing the transcript.
CRISPASR_NGRAM_LOOPFIX_OFF1 disables the repeated-n-gram collapse entirely, exposing the RAW decoded text. Diagnostic: for telling whether a loop originates in the decode itself or is merely being masked.

Decoding / beam search (shared)

VariablePurpose
CRISPASR_MAES_BETA / _MAES_GAMMA / _MAES_NUM_STEPSMAES beam-search parameters.
CRISPASR_TDT_BATCH / CRISPASR_RNNT_BATCHBatch the TDT / RNNT joint decode.
CRISPASR_RNNT_GGML_PERSTEPPer-step (vs. persistent-graph) ggml RNNT decode.
CRISPASR_NGRAM_LOOPFIX_OFFDisable the n-gram decode-loop breaker.
CRISPASR_GAP_FILL / _GAP_FILL_MIN_CSRe-transcribe spans a first pass left empty (long audio); on by default for parakeet, threshold non-JA 300 cs / JA 100 cs.

G2P / phonemizer

VariablePurpose
CRISPASR_CMUDICT_PATHPath to the CMUdict pronunciation dictionary.
CRISPASR_DE_DICT_PATH / _FR_DICT_PATH / _ES_DICT_PATHLanguage-specific pronunciation dictionaries.
CRISPASR_G2P_DICT_SOURCE / _G2P_MODEL_PATHG2P dictionary source / neural G2P model path.
CRISPASR_ESPEAK_DATA_PATHeSpeak-NG data directory.
CRISPASR_KOKORO_G2PKokoro G2P backend selection.
CRISPASR_KOKORO_MISAKI_IPA0 disables the espeak-IPA → misaki-alphabet conversion Kokoro needs (#316), restoring the raw G2P spelling for A/B. On by default.
CRISPASR_G2P_DE_UNSTRESS1 reads the German closed class the way espeak reads it in a SENTENCE (sie → ziː) instead of the citation form our per-word dictionary stores (zˈiː). Off by default: it takes phoneme agreement with espeak from 45.9% to 87.1%, but the ASR round-trip could not resolve a difference, and that metric measures intelligibility rather than naturalness (#316).
CRISPASR_KOKORO_DE_MISAKI_ALPHABET1 applies misaki's tied-sequence collapse for German (tsvˈaɪ → ʦvˈI), which is what the published training recipe does. Off by default: it made the ASR round-trip worse on the kokoro-de-hui-base model we ship, which appears to predate that part of the recipe (#316). The ʏ→y vocabulary fixup is applied either way.
CRISPASR_T5_REPEAT_BREAK0 disables the decode-loop break for madlad/T5 translation, restoring exact PyTorch-blueprint behaviour. On by default: MADLAD greedy-decodes into a repeated token cycle on some short inputs and burns the whole token budget on it. The blueprint does the same — this is a deliberate improvement on it, not a parity fix (#333).
CRISPASR_T5_NO_KV_REUSE1 re-forwards the whole decoder prefix each step instead of appending to the KV cache. Same output, much slower; an A/B lever for isolating cache bugs (#333).
CRISPASR_T5_KEEP_EMBED1 keeps shared.embed.* and lm_head.* at source precision when quantizing a T5 model. Off by default because it was measured and loses: on madlad400 it makes q8_0 3.38→3.62 GB and q4_k 2.04→2.41 GB for a worst-stage cosine that does not improve (0.999922→0.999920, 0.992929→0.992606). The Q4_K loss accumulates through the 32 encoder blocks, not in the embedding lookup (#333).
CRISPASR_KOKORO_PUNCT0 drops punctuation from the phoneme string for the German/French/Spanish built-in G2Ps, restoring pre-0.8.26 behaviour for A/B. On by default: Kokoro's vocabulary contains ,.;:!? and they are how it pauses (#316). English is not gated — it is settled against misaki.

Watermark / provenance

VariablePurpose
CRISPASR_NO_WATERMARKDisable the audio watermark.
CRISPASR_WATERMARK_LEGACYUse the legacy watermark path.
CRISPASR_NO_C2PA_REMUXSkip the C2PA MP4 remux step.

Quantization / diff-harness / misc

VariablePurpose
CRISPASR_QUANT_LMHEADQuantize the LM head during crispasr-quantize.
CRISPASR_IMATRIX_OUTImportance-matrix output path.
CRISPASR_ACTDUMP_OUT / _ACTDUMP_TENSORActivation dump output / target tensor.
CRISPASR_DIFF_NO_GPU / _DIFF_USE_GPU / _DIFF_SLICES / _DIFF_STAGEScrispasr-diff harness controls.
CRISPASR_MEL_PARALLEL / _MEL_TIMINGParallelize / time mel-spectrogram computation.
CRISPASR_VERBOSEGlobal verbose output.
CRISPASR_NO_WARMUP / CRISPASR_WARMUPSkip / force the model warmup pass.
CRISPASR_SERVER_WORKERS / CRISPASR_API_KEYSHTTP server worker count / API keys.

Reference-voice cache (voice cloning)

Encoding a reference clip for voice cloning (a codec encoder, a Conformer / Perceiver, or an ASR pass) is slow and produces a small, reusable blob. Every voice-cloning TTS backend caches that blob through one shared mechanism (src/core/tts_ref_cache.h), so the location and disable switch are identical across backends (issue #265 unified OmniVoice — which previously used a bespoke ~/.cache/crispasr cache — onto this path):

VariablePurpose
CRISPASR_TTS_REF_CACHESet to 0 to disable reference-voice caching everywhere.
CRISPASR_TTS_REF_CACHE_DIROverride the cache directory (default <TMPDIR>/crispasr-tts-refcache).

Content-addressed entries are keyed by a hash of the raw reference (plus, for some backends, an encoder-weight fingerprint) and tagged per backend (irodori-latent, openvoice2-se, omnivoice-voice, f5-reftext, …) so no two backends read each other's blob. OmniVoice additionally honors the legacy CRISPASR_OMNIVOICE_VOICE_CACHE=0 as an alias for the shared disable switch.

Variables CrispASR does not own

These are read by CrispASR but are OS / third-party conventions and are intentionally not renamed to the CRISPASR_ prefix:

VariableOrigin
HOME, USERPROFILE, LOCALAPPDATA, XDG_CACHE_HOMEOS home / cache dirs.
TMPDIR, TEMP, TMPOS temp dir (also the default ref-cache root).
HF_TOKEN, HUGGING_FACE_HUB_TOKENHuggingFace auth (Hub convention).
CUDA_VISIBLE_DEVICESCUDA device selection.
GGML_VK_VISIBLE_DEVICESggml Vulkan device selection.
LLAMA_*Vendored llama.cpp (talk-llama example).

Sibling modular libraries (own conventions)

The in-tree modular libraries keep their own established prefixes and are synced from their sibling repos, so their variables are not part of the CRISPASR_ scheme:

  • crisp_audio/ — audio tower: CRISP_AUDIO_DUMP_STAGES, CRISP_AUDIO_KEEP_PAD_FRAMES, CRISP_AUDIO_WINDOWED_ATTN.
  • glint/ — clean-room MP3/AAC codec: GLINT_*, AACDBG.
  • crisp_lid/ · crisp_punc/ · crisp_truecase/ — standalone LID / punctuation / truecasing libraries; they and their src/ counterparts follow each library's own naming (LID_*, FIREREDPUNC_*, PCS_*, TRUECASER_*). FIREREDPUNC_DEBUG=1 prints each restore pass as [PUNCDBG] in=<…> / out=<…>, which is the quickest way to see a backend's true model output — --no-punctuation is not, because it strips punctuation after the fact and so hides text the model punctuated itself.

    ⚠ These libraries are built by the main target, and each has a second copy under src/. src/CMakeLists.txt prefers crisp_punc/ (etc.) and falls back to the src/ copy only when the sibling directory is missing from a checkout — so the crisp_punc/ copy is what normally links, and a change must be applied to both. #308's capitalisation fix went into src/fireredpunc.cpp alone and was dead code for months while the shipping copy kept the bug. tests/test-punc-copies-in-sync.cpp now fails if they diverge.

Test fixtures

The live/integration test suite reads a few CRISPASR_-prefixed fixture variables (model/audio paths); tests/env-live-tests.sh sets sensible defaults. Like every other variable here, the pre-standardization bare names are still honored as deprecated aliases:

VariablePurpose
CRISPASR_PARAFORMER_MODEL / _PARAFORMER_MODEL_Q4K / _PARAFORMER_AUDIO_ZH / _PARAFORMER_AUDIO_ENParaformer test model / audio paths.
CRISPASR_AUDIOSEAL_GGUFAudioSeal test model path.
CRISPASR_PIPER_TEST_MODELPiper phonemize test model path.
CRISPASR_OV2_DUMP_DIROpenVoice2 HiFi test dump directory.
CRISPASR_MODEL_BTC_CHORDSBTC chord-recognition test model path. Defaults to $CRISPASR_MODELS_DIR/btc-chords-large-f32.gguf.

Per-backend variables

Every variable below is the canonical CRISPASR_-prefixed name; the bare form (without CRISPASR_) is accepted as a deprecated alias. See the suffix legend above for the meaning of the common _BENCH / _DEBUG / _DUMP_* / _GPU / _CPU / _FORCE_SCALAR / _FASTCONV / _CFG_INTERVAL / _SEED suffixes.

AAC codec

  • CRISPASR_AAC_DEBUG
  • CRISPASR_AAC_DECODER

ARK-ASR

  • CRISPASR_ARKASR_CPU
  • CRISPASR_ARKASR_DEBUG_GEN
  • CRISPASR_ARKASR_GPU
  • CRISPASR_ARKASR_MAX_SINGLE_PASS_S
  • CRISPASR_ARKASR_NO_CHUNK_CONTEXT
  • CRISPASR_ARKASR_NO_EOS_SUPPRESS
  • CRISPASR_ARKASR_TIMING

AudioSeal watermark

  • CRISPASR_AUDIOSEAL_BENCH
  • CRISPASR_AUDIOSEAL_DEBUG
  • CRISPASR_AUDIOSEAL_DUMP_STAGES

BananaMind TTS

  • CRISPASR_BANANAMIND_DEBUG
  • CRISPASR_BANANAMIND_TTS_BENCH

Bark

  • CRISPASR_BARK_BENCH
  • CRISPASR_BARK_DECODE_CODES
  • CRISPASR_BARK_DUMP_DIR

BERT encoder

  • CRISPASR_BERT_ENCODER_BENCH

BTC chord recognition

  • CRISPASR_BTC_MAJ_MIN — collapse the 170-class chord output to the 25-class maj/min vocabulary. Default off (full 170-class output): 170 reduces to maj/min at runtime, but a 25-class model can never be expanded, so the 170-class checkpoint is the shipping default.

The shipped BTC weights are CC-BY-NC-SA (trained on Isophonics / Robbie Williams / UsPop2002 chord annotations) even though the upstream BTC code and CrispASR itself are MIT. The registry refuses to download them without --accept-license cc-by-nc-sa-4.0 (or CRISPASR_ACCEPT_LICENSE). A commercial product must supply its own weights.

Canary / Canary-CTC / Canary-Qwen

  • CRISPASR_CANARY_BENCH
  • CRISPASR_CANARY_CTC_BENCH
  • CRISPASR_CANARY_QWEN_BENCH
  • CRISPASR_CANARY_QWEN_DEBUG
  • CRISPASR_CANARY_QWEN_MIN_ENC_FRAMES
  • CRISPASR_CANARY_QWEN_NO_ECHO_STRIP
  • CRISPASR_CANARY_LEGACY_STREAM
  • CRISPASR_CANARY_SEAM_DEDUP
  • CRISPASR_CANARY_STREAM_THRESHOLD_S

Chatterbox

  • CRISPASR_CHATTERBOX_BENCH
  • CRISPASR_CHATTERBOX_BENCH_B2
  • CRISPASR_CHATTERBOX_DEBUG
  • CRISPASR_CHATTERBOX_DUMP_ATTN_AT
  • CRISPASR_CHATTERBOX_DUMP_FFN_AT
  • CRISPASR_CHATTERBOX_DUMP_GPT2_LAYERS
  • CRISPASR_CHATTERBOX_DUMP_KPROJ_AT
  • CRISPASR_CHATTERBOX_DUMP_KROPE_AT
  • CRISPASR_CHATTERBOX_DUMP_KV_AT
  • CRISPASR_CHATTERBOX_DUMP_KV_LAYER
  • CRISPASR_CHATTERBOX_DUMP_LAYER
  • CRISPASR_CHATTERBOX_DUMP_LOGITS_AT
  • CRISPASR_CHATTERBOX_DUMP_NORM_AT
  • CRISPASR_CHATTERBOX_DUMP_QPROJ_AT
  • CRISPASR_CHATTERBOX_DUMP_VPROJ_AT
  • CRISPASR_CHATTERBOX_DUMP_WK
  • CRISPASR_CHATTERBOX_LANG
  • CRISPASR_CHATTERBOX_NAIVE_ATTN
  • CRISPASR_CHATTERBOX_SEED
  • CRISPASR_CHATTERBOX_SYN_TEXT
  • CRISPASR_CHATTERBOX_T3_BUCKET_REUSE
  • CRISPASR_CHATTERBOX_T3_CFG_B2
  • CRISPASR_CHATTERBOX_T3_CFG_BUCKET
  • CRISPASR_CHATTERBOX_T3_SEED
  • CRISPASR_CHATTERBOX_TEMP
  • CRISPASR_CHATTERBOX_THREADS

Chatterbox S3Gen

  • CRISPASR_S3GEN_CFG_INTERVAL
  • CRISPASR_S3GEN_DUMP
  • CRISPASR_S3GEN_DUMP_UNET
  • CRISPASR_S3GEN_DUMP_UNET_NO_AUTO_MARK
  • CRISPASR_S3GEN_FASTCONV
  • CRISPASR_S3GEN_FASTCONV_DEBUG
  • CRISPASR_S3GEN_RC_AS_MUL_MAT
  • CRISPASR_S3GEN_UNET_CFG_SINGLE
  • CRISPASR_S3GEN_UNET_CPU
  • CRISPASR_S3GEN_UNET_GALLOCR
  • CRISPASR_S3GEN_UNET_KEEP_GPU_OP
  • CRISPASR_S3GEN_UNET_MARK_DB_OUT
  • CRISPASR_S3GEN_UNET_MARK_DB_RESNET
  • CRISPASR_S3GEN_UNET_MARK_DB_TB
  • CRISPASR_S3GEN_UNET_MARK_MB_OUT
  • CRISPASR_S3GEN_UNET_MARK_MB_OUT_INDEX
  • CRISPASR_S3GEN_UNET_MARK_MB_OUT_MAX
  • CRISPASR_S3GEN_UNET_MARK_MB_RESNET
  • CRISPASR_S3GEN_UNET_PIN_CPU_OP
  • CRISPASR_S3GEN_UNET_PRESERVE_INTERMEDIATES
  • CRISPASR_S3GEN_UNET_PROBE_BLOCK1
  • CRISPASR_S3GEN_UNET_PROBE_DENOISER_OUT
  • CRISPASR_S3GEN_UNET_PROBE_INPUT_SNAPSHOT
  • CRISPASR_S3GEN_UNET_PROBE_RC_OUT

Chatterbox sub-modules

  • CRISPASR_CB_CAMPPLUS_BENCH
  • CRISPASR_CB_S3GEN_BENCH
  • CRISPASR_CB_S3TOK_BENCH
  • CRISPASR_CB_VE_BENCH

Cohere

  • CRISPASR_COHERE_BENCH
  • CRISPASR_COHERE_DEBUG
  • CRISPASR_COHERE_DEVICE
  • CRISPASR_COHERE_DUMP_ATTN
  • CRISPASR_COHERE_DUMP_ENCOUT
  • CRISPASR_COHERE_DUMP_MEL
  • CRISPASR_COHERE_DUMP_STAGES
  • CRISPASR_COHERE_FLASH
  • CRISPASR_COHERE_GAPS
  • CRISPASR_COHERE_LEGACY_SA
  • CRISPASR_COHERE_PROF
  • CRISPASR_COHERE_THREADS

CosyVoice3

  • CRISPASR_COSYVOICE3_BENCH
  • CRISPASR_COSYVOICE3_CAMPPLUS_PATH
  • CRISPASR_COSYVOICE3_CFG_BATCH
  • CRISPASR_COSYVOICE3_CFG_INTERVAL
  • CRISPASR_COSYVOICE3_CFG_INTERVAL_DEBUG
  • CRISPASR_COSYVOICE3_DUMP_TOKENS
  • CRISPASR_COSYVOICE3_FASTCONV
  • CRISPASR_COSYVOICE3_FASTCONV_DEBUG
  • CRISPASR_COSYVOICE3_FLOW_STEPS
  • CRISPASR_COSYVOICE3_HIFT_PATH
  • CRISPASR_COSYVOICE3_KV_BUCKET
  • CRISPASR_COSYVOICE3_NO_CLONE_CACHE — re-extract the --voice ref.wav speaker (s3tokenizer + CAMPPlus + prompt mel) on every synthesis instead of once per reference. Output-identical; the cached path is ~30% faster on a multi-sentence --tts (#334).
  • CRISPASR_COSYVOICE3_NO_MIN_LEN — drop the decode's minimum-length floor (2 speech tokens per target text token, upstream's min_token_text_ratio). Without the floor a single unlucky sample at step 0 ends the decode with no audio at all (#334).
  • CRISPASR_COSYVOICE3_VOICES_PATH

CosyVoice3 (diff-harness assets)

  • CRISPASR_CV3_FLOW_GGUF
  • CRISPASR_CV3_HIFT_GGUF
  • CRISPASR_CV3_S3TOK_GGUF

CSM TTS

  • CRISPASR_CSM_BENCH
  • CRISPASR_CSM_WAV_FRAMES
  • CRISPASR_CSM_WAV_OUT
  • CRISPASR_CSM_WAV_TEMP
  • CRISPASR_CSM_WAV_TEXT

Dia TTS

  • CRISPASR_DIA_BENCH
  • CRISPASR_DIA_DECODE_CODES
  • CRISPASR_DIA_DUMP_DIR
  • CRISPASR_DIA_DUMP_STEPLOGITS
  • CRISPASR_DIA_DUMP_TOKENS
  • CRISPASR_DIA_FORCE_TOKENS
  • CRISPASR_DIA_GREEDY
  • CRISPASR_DIA_MAX_STEPS
  • CRISPASR_DIA_TTS_GPU

dots.tts

  • CRISPASR_DOTS_CFG_INTERVAL
  • CRISPASR_DOTS_CFG_INTERVAL_DEBUG
  • CRISPASR_DOTS_DIFF_GPU
  • CRISPASR_DOTS_DIT_DEBUG
  • CRISPASR_DOTS_EOS_THRESHOLD
  • CRISPASR_DOTS_FM_AB
  • CRISPASR_DOTS_FM_DUMP
  • CRISPASR_DOTS_FUSED_STEP
  • CRISPASR_DOTS_MAX_PATCHES
  • CRISPASR_DOTS_ODE_STEPS
  • CRISPASR_DOTS_PENC_VERIFY
  • CRISPASR_DOTS_TTS_BENCH
  • CRISPASR_DOTS_TTS_CPU
  • CRISPASR_DOTS_TTS_DEBUG

ECAPA (LID / speaker)

  • CRISPASR_ECAPA_ASP_CPU
  • CRISPASR_ECAPA_ASP_GGML
  • CRISPASR_ECAPA_FORCE_SCALAR
  • CRISPASR_ECAPA_LID_BENCH
  • CRISPASR_ECAPA_REF_FBANK
  • CRISPASR_ECAPA_TIMING

F5-TTS

  • CRISPASR_F5_BATCH_CFG
  • CRISPASR_F5_BENCH
  • CRISPASR_F5_CFG_INTERVAL
  • CRISPASR_F5_DURATION_CLAMP — clamp the per-char speech rate into a sane English band so a reference whose audio/transcript lengths are mismatched can't truncate (or balloon) the output (#294). Default on; set 0 to restore the exact upstream ref_T / ref_text_len * gen_text_len / speed estimate.
  • CRISPASR_F5_FORCE_SCALAR
  • CRISPASR_F5_REF_MAX_SEC — clip the reference audio to this many seconds before it drives the duration estimate (upstream parity: 12 s). Default 12; set 0 to disable the clip.
  • CRISPASR_F5_REF_TRIM_SILENCE — strip leading/trailing silence and collapse internal silences >~1 s in the reference audio (upstream parity). Default on; set 0 to disable.

FastConformer (shared encoder)

  • CRISPASR_FC_ATTN_CONT
  • CRISPASR_FC_BUCKET
  • CRISPASR_FC_FUSED_QKV
  • CRISPASR_FC_GPU_MANUAL_ATTN
  • CRISPASR_FC_MAX_LAYERS
  • CRISPASR_FC_MEM_DEBUG
  • CRISPASR_FC_NO_FLASH
  • CRISPASR_FC_PROFILE
  • CRISPASR_FC_PROF_FP
  • CRISPASR_FC_PROF_FP_COLS
  • CRISPASR_FC_PW_Q8
  • CRISPASR_FC_TILED_ATTN
  • CRISPASR_FC_TILED_BLOCK
  • CRISPASR_FC_WINDOWED_ATTN
  • CRISPASR_FC_WINDOW_BLOCK

FastPitch

  • CRISPASR_FASTPITCH_BENCH
  • CRISPASR_FASTPITCH_DUMP_DIR
  • CRISPASR_FASTPITCH_FORCE_TOKENS

Ffmpeg

  • CRISPASR_FFMPEG_LOG

FireRed ASR / VAD

  • CRISPASR_FIRERED_BEAM_F32
  • CRISPASR_FIRERED_BENCH
  • CRISPASR_FIRERED_ENC_CPU
  • CRISPASR_FIRERED_GGML_ATTN
  • CRISPASR_FIRERED_LID_BENCH
  • CRISPASR_FIRERED_MATVEC_CACHE
  • CRISPASR_FIRERED_NO_REPEAT_BREAK
  • CRISPASR_FIRERED_VAD_BENCH
  • CRISPASR_FIRERED_VAD_DEBUG
  • CRISPASR_FIRERED_VAD_FORCE_SCALAR

FunASR / SenseVoice

  • CRISPASR_FUNASR_BENCH
  • CRISPASR_FUNASR_DUMP_STAGES
  • CRISPASR_FUNASR_EMBED_FAST
  • CRISPASR_FUNASR_LLM_CPU
  • CRISPASR_FUNASR_LLM_LAYERS
  • CRISPASR_FUNASR_NAN_CHECK
  • CRISPASR_FUNASR_NO_FA
  • CRISPASR_FUNASR_STEP_CACHE

Gemma-4 E2B

  • CRISPASR_GEMMA4_AUTO_CHUNK
  • CRISPASR_GEMMA4_E2B_BENCH
  • CRISPASR_GEMMA4_E2B_EMBED_FAST

GLM-ASR

  • CRISPASR_GLM_ASR_BENCH
  • CRISPASR_GLM_ASR_DEBUG
  • CRISPASR_GLM_ASR_EMBED_FAST
  • CRISPASR_GLM_ASR_LEGACY_PROMPT
  • CRISPASR_GLM_ASR_SINGLE_WINDOW

Granite speech / NLE

  • CRISPASR_GRANITE_BENCH
  • CRISPASR_GRANITE_DEC_GALLOCR
  • CRISPASR_GRANITE_DEC_PROFILE
  • CRISPASR_GRANITE_DISABLE_ENCODER_GRAPH
  • CRISPASR_GRANITE_ENC_F16
  • CRISPASR_GRANITE_FORCE_SCALAR
  • CRISPASR_GRANITE_NLE_BENCH
  • CRISPASR_GRANITE_NLE_EDIT_DUMP
  • CRISPASR_GRANITE_QUANT_ALL

HiFT vocoder

  • CRISPASR_HIFT_FULL_IDFT

HTDemucs (source separation)

  • CRISPASR_HTDEMUCS_BLAS — route the CrossTransformer matmuls through cblas_sgemm (default ON where Accelerate is available). =0 selects the scalar path. The transformer is ~86% of an unoptimised forward pass, so this is the dominant knob (measured 44x on the transformer, 4.6x overall).
  • CRISPASR_HTDEMUCS_FASTCONV — batched im2col + one GEMM for the CPU convs (default ON). =0 selects the original per-time-frame scalar path. Measured enc.conv2d 10.0 s -> 0.17 s and enc.rewrite 12.2 s -> 0.30 s.
  • CRISPASR_HTDEMUCS_WCACHE — cache F32 copies of weight tensors by pointer (default ON). =0 re-reads and re-converts on every access, which the DConv stacks do ~6k times per encoder layer.
  • CRISPASR_HTDEMUCS_GGML — run the CrossTransformer as a ggml graph instead of the CPU/BLAS path (default OFF). Verified correct on CPU and Metal (45/45 stages, every layer cos 1.000000) but not yet proven faster overall, so it stays opt-in per the inverse-default rule.
  • CRISPASR_HTDEMUCS_GPU — request a GPU backend (CUDA > Metal > Vulkan, CPU fallback). Only meaningful together with _GGML=1: under the CPU/BLAS path the weights would sit on the device and every kernel would pay a read back.
  • CRISPASR_HTDEMUCS_PROFILE — print a per-phase wall-time breakdown of one forward pass (stft / enc / transformer / dec / istft).
  • CRISPASR_HTDEMUCS_DEBUG — verbose per-layer shape and NaN diagnostics.
  • CRISPASR_HTDEMUCS_SKIP_TIME — skip the time branch (bisection aid).

All three optimisation gates are output-equivalent: the per-stage diff reports 45/45 stages passing with them ON or OFF.

Higgs STT

  • CRISPASR_HIGGS_DEBUG
  • CRISPASR_HIGGS_STT_BENCH
  • CRISPASR_HIGGS_STT_EMBED_FAST
  • CRISPASR_HIGGS_STT_FUSED_QKV

IndexTTS

  • CRISPASR_INDEXTTS_AA_BACKEND
  • CRISPASR_INDEXTTS_AA_SCALAR
  • CRISPASR_INDEXTTS_AUDIO24K_FILE
  • CRISPASR_INDEXTTS_BEAM_SIZE
  • CRISPASR_INDEXTTS_BENCH
  • CRISPASR_INDEXTTS_COND_FILE
  • CRISPASR_INDEXTTS_DEBUG
  • CRISPASR_INDEXTTS_KV_DEVICE_COPY
  • CRISPASR_INDEXTTS_LATENT_FILE
  • CRISPASR_INDEXTTS_MEL_CODES_FILE
  • CRISPASR_INDEXTTS_MEL_FILE
  • CRISPASR_INDEXTTS_SPK_NORM
  • CRISPASR_INDEXTTS_TEXT_NORMALIZER
  • CRISPASR_INDEXTTS_VOCODER_AA
  • CRISPASR_INDEXTTS_VOCODER_RAW
  • CRISPASR_INDEXTTS_VOC_BENCH
  • CRISPASR_INDEXTTS_VOC_FORCE_GPU

Irodori TTS

  • CRISPASR_IRODORI_CAPTION
  • CRISPASR_IRODORI_CAPTION_TOKEN_IDS
  • CRISPASR_IRODORI_CFG_CAPTION
  • CRISPASR_IRODORI_CFG_INTERVAL
  • CRISPASR_IRODORI_CFG_INTERVAL_DEBUG
  • CRISPASR_IRODORI_CFG_SPEAKER
  • CRISPASR_IRODORI_CFG_TEXT
  • CRISPASR_IRODORI_CODEC_CPU
  • CRISPASR_IRODORI_CODEC_GPU
  • CRISPASR_IRODORI_CPU
  • CRISPASR_IRODORI_DEBUG
  • CRISPASR_IRODORI_DECODE_CHUNK
  • CRISPASR_IRODORI_DECODE_CTX
  • CRISPASR_IRODORI_DIT_TIMING
  • CRISPASR_IRODORI_DUMP_LATENT
  • CRISPASR_IRODORI_DUMP_TEXT_STATE
  • CRISPASR_IRODORI_DUMP_TOKENS
  • CRISPASR_IRODORI_DUMP_V_PRED0
  • CRISPASR_IRODORI_ENC_DUMP
  • CRISPASR_IRODORI_ENC_PRENORM
  • CRISPASR_IRODORI_FASTCONV
  • CRISPASR_IRODORI_LAYERS
  • CRISPASR_IRODORI_ODE_STEPS
  • CRISPASR_IRODORI_PERSIST_GRAPH
  • CRISPASR_IRODORI_REF_NOISE
  • CRISPASR_IRODORI_TOKEN_IDS
  • CRISPASR_IRODORI_T_LATENT

Kokoro

  • CRISPASR_KOKORO_BENCH
  • CRISPASR_KOKORO_DEBUG
  • CRISPASR_KOKORO_DEBUG_INTERMEDIATES
  • CRISPASR_KOKORO_DUMP_STAGES
  • CRISPASR_KOKORO_FASTCONV
  • CRISPASR_KOKORO_FASTCONV_DEBUG
  • CRISPASR_KOKORO_G2P
  • CRISPASR_KOKORO_SEED
  • CRISPASR_KOKORO_USE_GPU
  • CRISPASR_KOKORO_VOICE_GGUF

KugelAudio

  • CRISPASR_KUGELAUDIO_CPU_ONLY
  • CRISPASR_KUGELAUDIO_DEBUG

Kyutai STT

  • CRISPASR_KYUTAI_RVQ_FAST
  • CRISPASR_KYUTAI_STT_BENCH

LFM2-Audio

  • CRISPASR_LFM2_AUDIO_BENCH
  • CRISPASR_LFM2_AUDIO_CPU
  • CRISPASR_LFM2_SNAP_LAYERS

M2M-100 translate

  • CRISPASR_M2M100_BENCH
  • CRISPASR_M2M100_GPU

MarbleNet VAD

  • CRISPASR_MARBLENET_VAD_BENCH

MeloTTS

  • CRISPASR_MELOTTS_BENCH
  • CRISPASR_MELOTTS_BERT
  • CRISPASR_MELOTTS_FORCE_SCALAR
  • CRISPASR_MELOTTS_WEIGHT_CACHE

Mimi codec

  • CRISPASR_MIMI_NONCAUSAL

MiMo-ASR

  • CRISPASR_MIMO_ASR_BENCH
  • CRISPASR_MIMO_ASR_DIAG
  • CRISPASR_MIMO_ASR_DUMP_STAGES
  • CRISPASR_MIMO_ASR_GPU
  • CRISPASR_MIMO_FORCE_CPU
  • CRISPASR_MIMO_SMOKE_DUMP
  • CRISPASR_MIMO_SMOKE_GPU
  • CRISPASR_MIMO_TOKENIZER_GPU
  • CRISPASR_MIMO_TOK_CPU

mini-omni2

  • CRISPASR_MINI_OMNI2_BENCH

Moonshine

  • CRISPASR_MOONSHINE_ALL_GPU
  • CRISPASR_MOONSHINE_BENCH
  • CRISPASR_MOONSHINE_ENC_ATTN
  • CRISPASR_MOONSHINE_NO_REPEAT_BREAK
  • CRISPASR_MOONSHINE_STREAMING_BENCH
  • CRISPASR_MOONSHINE_STREAMING_GPU
  • CRISPASR_MOONSHINE_STREAM_BENCH

MOSS family

  • CRISPASR_MOSS_AUDIO_BENCH
  • CRISPASR_MOSS_AUDIO_EMBED_FAST
  • CRISPASR_MOSS_AUDIO_ENC_FLASH
  • CRISPASR_MOSS_AUDIO_ENC_MANUAL
  • CRISPASR_MOSS_AUDIO_FORCE_CPU
  • CRISPASR_MOSS_AUDIO_MEL_FILE
  • CRISPASR_MOSS_DIARIZE_BENCH
  • CRISPASR_MOSS_DIARIZE_DEBUG
  • CRISPASR_MOSS_DIARIZE_DUMP_CONV
  • CRISPASR_MOSS_DIARIZE_ENC_FLASH
  • CRISPASR_MOSS_DIARIZE_ENC_MANUAL
  • CRISPASR_MOSS_DIARIZE_FORCE_CPU
  • CRISPASR_MOSS_DIARIZE_NO_LOOPFIX
  • CRISPASR_MOSS_TRANSCRIBE_BENCH
  • CRISPASR_MOSS_TRANSCRIBE_ENC_DUMP
  • CRISPASR_MOSS_TRANSCRIBE_ENC_FLASH
  • CRISPASR_MOSS_TRANSCRIBE_ENC_MANUAL
  • CRISPASR_MOSS_TRANSCRIBE_FORCE_CPU
  • CRISPASR_MOSS_TRANSCRIBE_L0_DUMP
  • CRISPASR_MOSS_TRANSCRIBE_MEL_DUMP
  • CRISPASR_MOSS_TRANSCRIBE_NO_LOOPFIX
  • CRISPASR_MOSS_TTS_BENCH
  • CRISPASR_MOSS_TTS_LOCAL_DEBUG
  • CRISPASR_MOSS_TTS_LOCAL_GREEDY_AUDIO
  • CRISPASR_MOSS_TTS_LOCAL_NO_GPU

MP3 codec

  • CRISPASR_MP3_ENCODER

Diarization — foxnose (#324)

  • CRISPASR_DIARIZE_COUNT — speaker-count estimator: bic (default, the upstream GMM/BIC + silhouette sweep) or eigengap. Eigengap is better on well-separated synthetic data and cheaper, but under-counts on real speech (11.4 % vs 5.3 % DER on VoxConverse) — see docs/foxnose-diarize/PLAN.md
  • CRISPASR_DIARIZE_BIC_WINDOW — score silhouette only in a [k-2, k+3] window around the BIC anchor instead of the full [min, max] range. The full range is the default: the BIC anchor is unreliable in both directions (measured errors of +5 / -3 / -3 on 4/5/6 well-separated blobs) and when it over-counts the window is stranded above the truth and cannot climb back to it
  • CRISPASR_WESPEAKER_BENCH — per-stage embedder timings (fbank / resnet / resnet_windows). Counting invocations of these is also how you check WHICH embedding path actually ran
  • CRISPASR_WESPEAKER_DEBUG — embedder diagnostics
  • CRISPASR_DIARIZE_DEBUG — chosen speaker count, the reason it was chosen, and the per-k silhouette curve behind it. Worth reading before trusting a count: on a borderline file the decision can rest on a <1 % score gap
  • CRISPASR_DIARIZE_EMBED_WORKERS — windows embedded concurrently (default: -t). Each worker gets its own context sharing one copy of the weights
  • CRISPASR_SPEAKER_EMBED_THREADS — ggml threads per embedder context (default: -t). Honoured by the pluggable embedders and by wespeaker
  • CRISPASR_DIARIZE_SPAN_EMBED=1 — run ONE network pass per span of windows instead of one per window. 1.78x less diarization CPU for +0.30 mean DER on the VoxConverse dev shard; off by default because accuracy is the better default for a diarizer. See docs/cli.md#diarization
  • CRISPASR_DIARIZE_SPAN_WINDOWS — windows per span (default 32). Measured NOT to affect the accuracy cost — identical from N=2 to N=32 — so there is nothing to tune here; larger is simply faster

GigaAM-v3

  • CRISPASR_GIGAAM_BENCH — per-stage timings (mel / encoder / decode)
  • CRISPASR_GIGAAM_DEBUG — encoder output min/max/mean
  • CRISPASR_GIGAAM_FLASH — ggml_flash_attn_ext in the encoder (opt-in; the manual QK^T path is what the per-stage diff was validated on)
  • CRISPASR_GIGAAM_FORCE_SCALAR — scalar LSTM/joint loops instead of cblas
  • CRISPASR_GIGAAM_QUANT_ALL — let crispasr-quantize quantize the heads and the pre-encode convs too (default keeps them at source precision)

Nemotron

  • CRISPASR_NEMOTRON_BENCH
  • CRISPASR_NEMOTRON_CONTEXT_PRESET
  • CRISPASR_NEMOTRON_DEBUG
  • CRISPASR_NEMOTRON_DECODE_TIMING
  • CRISPASR_NEMOTRON_FORCE_SCALAR
  • CRISPASR_NEMOTRON_GGML_DECODE
  • CRISPASR_NEMOTRON_MAES
  • CRISPASR_NEMOTRON_NO_WINDOW_MASK
  • CRISPASR_NEMOTRON_STREAMING

OmniASR

  • CRISPASR_OMNIASR_BENCH
  • CRISPASR_OMNIASR_DEBUG
  • CRISPASR_OMNIASR_DUMP_DIR
  • CRISPASR_OMNIASR_KEEP_F16_HEAD
  • CRISPASR_OMNIASR_KEEP_F16_TAIL
  • CRISPASR_OMNIASR_QUANT_ALL

OmniVoice

  • CRISPASR_OMNIVOICE_ACENC_BISECT
  • CRISPASR_OMNIVOICE_AUTO_LANG — default on. When no language was requested (-l / -tl / the server's "language" / set_target_language), guess one from the text being spoken and use it if it maps to an id the model knows. An explicitly requested language always wins; a low-confidence guess resolves to nothing and behaves exactly as before. =0 restores the old always-language-agnostic behaviour. Exists because SubtitleEdit's language menu is not yet wired to its request payload (#13273).
  • CRISPASR_OMNIVOICE_BENCH
  • CRISPASR_OMNIVOICE_CFG_INTERVAL
  • CRISPASR_OMNIVOICE_CHUNK
  • CRISPASR_OMNIVOICE_CLASS_TEMP
  • CRISPASR_OMNIVOICE_CODEC_FASTCONV
  • CRISPASR_OMNIVOICE_CODEC_GPU — codec placement override (1 = GPU, 0 = CPU). Unset defaults to GPU on CUDA and CPU on Metal/CPU.
  • CRISPASR_OMNIVOICE_CPU
  • CRISPASR_OMNIVOICE_DEBUG_CODES
  • CRISPASR_OMNIVOICE_DEBUG_SUM
  • CRISPASR_OMNIVOICE_ENCODE_DIFF
  • CRISPASR_OMNIVOICE_FRAMES_PER_CHAR
  • CRISPASR_OMNIVOICE_FUSED_STEP
  • CRISPASR_OMNIVOICE_GUIDANCE
  • CRISPASR_OMNIVOICE_HUBERT_REF
  • CRISPASR_OMNIVOICE_NUM_STEPS
  • CRISPASR_OMNIVOICE_PERSISTENT_GRAPH
  • CRISPASR_OMNIVOICE_POS_TEMP
  • CRISPASR_OMNIVOICE_UNIFIED_CFG
  • CRISPASR_OMNIVOICE_VOICE_CACHE

OpenVoice2

  • CRISPASR_OPENVOICE2_BENCH

OpenVoice2

  • CRISPASR_OV2_DUMP_DIR
  • CRISPASR_OV2_FORCE_SCALAR
  • CRISPASR_OV2_NO_NORMALIZE
  • CRISPASR_OV2_TAU

Opus codec

  • CRISPASR_OPUS_DEBUG
  • CRISPASR_OPUS_DECODER
  • CRISPASR_OPUS_ENCODER

Orpheus

  • CRISPASR_ORPHEUS_BENCH
  • CRISPASR_ORPHEUS_BUCKET
  • CRISPASR_ORPHEUS_DEBUG
  • CRISPASR_ORPHEUS_DIFF_GPU
  • CRISPASR_ORPHEUS_DIFF_MAXGEN
  • CRISPASR_ORPHEUS_PROMPT_IDS
  • CRISPASR_ORPHEUS_SNAC_CODE
  • CRISPASR_ORPHEUS_SNAC_GPU
  • CRISPASR_ORPHEUS_SNAC_T_SUPER

OuteTTS

  • CRISPASR_OUTETTS_BENCH

Paraformer

  • CRISPASR_PARAFORMER_BENCH
  • CRISPASR_PARAFORMER_GPU

Parakeet

  • CRISPASR_PARAKEET_ATT_CONTEXT
  • CRISPASR_PARAKEET_BENCH
  • CRISPASR_PARAKEET_CHUNK_OVERLAP
  • CRISPASR_PARAKEET_CHUNK_SECONDS
  • CRISPASR_PARAKEET_DEBUG
  • CRISPASR_PARAKEET_DECODE_TIMING
  • CRISPASR_PARAKEET_ENC_CACHE
  • CRISPASR_PARAKEET_ENC_PROBE
  • CRISPASR_PARAKEET_FORCE_SCALAR
  • CRISPASR_PARAKEET_GGML_DECODE
  • CRISPASR_PARAKEET_INTERNAL_CHUNKING
  • CRISPASR_PARAKEET_LONGFORM
  • CRISPASR_PARAKEET_MAES
  • CRISPASR_PARAKEET_MEM_COEFF
  • CRISPASR_PARAKEET_MEM_POLICY
  • CRISPASR_PARAKEET_QUANT_ALL
  • CRISPASR_PARAKEET_SIMULATE_ENCODE_OOM
  • CRISPASR_PARAKEET_STREAM_CHUNK
  • CRISPASR_PARAKEET_STREAM_OVERLAP
  • CRISPASR_PARAKEET_STREAM_THRESHOLD
  • CRISPASR_PARAKEET_VAD_SLICE_CAP
  • CRISPASR_PARAKEET_VRAM_BUDGET_MB

Parler-TTS

  • CRISPASR_PARLER_BUCKET
  • CRISPASR_PARLER_DEBUG
  • CRISPASR_PARLER_DESC_IDS
  • CRISPASR_PARLER_DIFF_MAXGEN
  • CRISPASR_PARLER_DUMP_ENC
  • CRISPASR_PARLER_PROMPT_IDS
  • CRISPASR_PARLER_TTS_BENCH

Piper

  • CRISPASR_PIPER_FORCE_SCALAR
  • CRISPASR_PIPER_TTS_BENCH
  • CRISPASR_PIPER_WEIGHT_CACHE

Pocket-TTS

  • CRISPASR_POCKET_DUMP_DIR
  • CRISPASR_POCKET_FORCE_LATENTS
  • CRISPASR_POCKET_FORCE_NOISE
  • CRISPASR_POCKET_MANUAL_BACKBONE
  • CRISPASR_POCKET_MANUAL_MIMI
  • CRISPASR_POCKET_MAX_FRAMES
  • CRISPASR_POCKET_MIMI_DUMP
  • CRISPASR_POCKET_MIMI_SCALAR
  • CRISPASR_POCKET_TTS_BENCH
  • CRISPASR_POCKET_VOICE_CACHE
  • CRISPASR_POCKET_VULKAN_MIMI_MAX_FRAMES

Pyannote segmentation

  • CRISPASR_PYANNOTE_LEGACY
  • CRISPASR_PYANNOTE_SEG_BENCH
  • CRISPASR_PYANNOTE_SEG_DUMP
  • CRISPASR_PYANNOTE_CHUNK_S — audio per chunk of parallel inference, in seconds (default 60; 0 restores the pre-#326 single scan over the whole file). Chunking is decided by audio LENGTH and never by thread count, so posteriors do not change with -t
  • CRISPASR_PYANNOTE_CHUNK_CONTEXT_S — real audio spliced either side of a chunk and then trimmed (default 5), which absorbs the convolutions' zero padding and the LSTM's zero initial state

Qwen3-ASR

  • CRISPASR_QWEN3ASR_QUANT_AUDIO

Qwen3-ASR / Qwen3-TTS

  • CRISPASR_QWEN3_ASR_BENCH
  • CRISPASR_QWEN3_ASR_EMBED_FAST
  • CRISPASR_QWEN3_ASR_FUSED_QKV
  • CRISPASR_QWEN3_SYSPROMPT_LANG
  • CRISPASR_QWEN3_TTS_BENCH
  • CRISPASR_QWEN3_TTS_CODEC_ALLOW_FULL
  • CRISPASR_QWEN3_TTS_CODEC_CHUNK
  • CRISPASR_QWEN3_TTS_CODEC_CPU
  • CRISPASR_QWEN3_TTS_CODEC_CTX
  • CRISPASR_QWEN3_TTS_CODEC_FORCE_METAL
  • CRISPASR_QWEN3_TTS_CODEC_GGUF
  • CRISPASR_QWEN3_TTS_CODEC_GPU
  • CRISPASR_QWEN3_TTS_CODEC_TRACE
  • CRISPASR_QWEN3_TTS_EMBD_CHECK
  • CRISPASR_QWEN3_TTS_DUMP_LOGITS=<dir> — write the raw per-frame talker logits (f32, before the repetition penalty and the suppress mask) plus a top-5 line to stderr. The instrument for a cross-backend diff: tokens alone cannot tell a miscompute from amplified rounding (#337).
  • CRISPASR_QWEN3_TTS_REPLAY_CODES=<file> — 16 whitespace-separated codec ids per frame; the decode uses them instead of sampling. Teacher forcing, and the ONLY way to compare two backends step by step: pin the whole frame or the 15 residual codebooks (which must be sampled) diverge and the diff measures trajectory, not arithmetic (#337).
  • CRISPASR_QWEN3_TTS_REPLAY_TOKENS=<file> — the weaker form: codebook-0 ids only. Useful for forcing a trajectory, NOT sufficient for a logits diff.
  • CRISPASR_QWEN3_TTS_GREEDY — force the talker's codebook-0 sampler to argmax (top_k=1). The frame sequence then depends only on the logits, so two backends agree if and only if their logits agree — this is the lever for telling a GPU miscompute apart from a sampling difference, and without it a CPU-vs-GPU token comparison proves nothing (#337).
  • CRISPASR_QWEN3_TTS_MAX_FRAMES
  • CRISPASR_QWEN3_TTS_SKIP_REF_DECODE

SenseVoice

  • CRISPASR_SENSEVOICE_BENCH
  • CRISPASR_SENSEVOICE_NO_FA

Sidon

  • CRISPASR_SIDON_FASTCONV — DAC convolution mode (off, k1-f16, k1-f32, or full). Unset defaults to k1-f16 on CUDA and off on Vulkan/CPU.
  • CRISPASR_SIDON_RPE — relative-position-bias formulation: bucket-direct (default), bucket, or expand (legacy [head_dim, T, T] expansion, ~1 GiB more predictor workspace at T≈2825; keeps the Vulkan mul_mat batching branch). All three are algebraically equivalent.
  • CRISPASR_SIDON_DECODER_CHUNK_FRAMES — maximum DAC core size in feature frames (default 512). 0 decodes the whole utterance in one graph (~4.5 GiB at T≈2825 vs ~0.79 GiB chunked). Chunked output is bit-exact against the whole-utterance decode.
  • CRISPASR_SIDON_LOOKAHEAD — set to 0 to disable the input padding (one leading predictor frame plus 1.5 s of right-side lookahead). Padding is on by default; without it the last ~12 ms of every clip is a full-scale transient.
  • CRISPASR_SIDON_MAX_FRAMES — predictor input cap in feature frames (default 3000, ~58.5 s after the lookahead). Guards the O(T^2) attention.
  • CRISPASR_SIDON_DEBUG — print per-stage scheduler workspace sizes (per backend) after graph allocation.

Sherpa

  • CRISPASR_SHERPA_LID_BIN

Silero LID

  • CRISPASR_SILERO_FORCE_SCALAR
  • CRISPASR_SILERO_LID_BENCH
  • CRISPASR_SILERO_LID_DEBUG
  • CRISPASR_SILERO_LID_DUMP
  • CRISPASR_SILERO_LID_LEGACY
  • CRISPASR_SILERO_LID_MAX_S
  • CRISPASR_SILERO_LID_TRACE
  • CRISPASR_SILERO_LID_TRACE_OFF
  • CRISPASR_SILERO_LID_TRUNC
  • CRISPASR_SILERO_LID_VULKAN

SpeechT5

  • CRISPASR_SPEECHT5_DUMP_DIR
  • CRISPASR_SPEECHT5_FASTCONV
  • CRISPASR_SPEECHT5_FASTCONV_DEBUG
  • CRISPASR_SPEECHT5_TTS_BENCH

T5 translate

  • CRISPASR_T5_GPU
  • CRISPASR_T5_TRANSLATE_BENCH

TaDa TTS

  • CRISPASR_TADA_ACOUSTIC_CFG
  • CRISPASR_TADA_ALLOW_VULKAN
  • CRISPASR_TADA_BATCH_PREFILL
  • CRISPASR_TADA_BENCH
  • CRISPASR_TADA_BUCKET_MIN
  • CRISPASR_TADA_CFG_INTERVAL
  • CRISPASR_TADA_CFG_INTERVAL_DEBUG
  • CRISPASR_TADA_CODEC_BENCH
  • CRISPASR_TADA_CODEC_DUMP
  • CRISPASR_TADA_CODEC_GGUF
  • CRISPASR_TADA_CODEC_VULKAN_NATIVE
  • CRISPASR_TADA_CTC_ASR
  • CRISPASR_TADA_DIFF_TEXT
  • CRISPASR_TADA_DO_SAMPLE
  • CRISPASR_TADA_DUMP_ACOUSTIC_FEATURES
  • CRISPASR_TADA_DUMP_FEATURES
  • CRISPASR_TADA_DUMP_FM_STEPS
  • CRISPASR_TADA_DUMP_TIME_BEFORE
  • CRISPASR_TADA_ENCODER_DEBUG
  • CRISPASR_TADA_EXTRA_STEPS
  • CRISPASR_TADA_FM_B2
  • CRISPASR_TADA_KEEP_F16_HEAD
  • CRISPASR_TADA_KEEP_F16_TAIL
  • CRISPASR_TADA_MAX_EXPANDED_FRAMES
  • CRISPASR_TADA_NOISE_TEMP
  • CRISPASR_TADA_NO_BUCKET
  • CRISPASR_TADA_NUM_CANDIDATES
  • CRISPASR_TADA_NUM_FM_STEPS
  • CRISPASR_TADA_PROMPT_CACHE
  • CRISPASR_TADA_PROMPT_TEXT
  • CRISPASR_TADA_QUANT_ALL
  • CRISPASR_TADA_REPETITION_PENALTY
  • CRISPASR_TADA_SCORER
  • CRISPASR_TADA_TALKER_TIMING
  • CRISPASR_TADA_TEMPERATURE
  • CRISPASR_TADA_TOP_K
  • CRISPASR_TADA_TOP_P
  • CRISPASR_TADA_VULKAN_NATIVE

TitaNet speaker

  • CRISPASR_TITANET_BENCH
  • CRISPASR_TITANET_DUMP
  • CRISPASR_TITANET_DUMP_MEL
  • CRISPASR_TITANET_FORCE_SCALAR
  • CRISPASR_TITANET_GGML
  • CRISPASR_TITANET_GPU
  • CRISPASR_TITANET_LEGACY
  • CRISPASR_TITANET_REF_MEL

Three compute paths exist and all are kept working; the default is the fastest one measured per platform.

pathselected bymeasured, M1, 2 s segment
legacy (Accelerate / hand-rolled)default where HAVE_ACCELERATE71.7 ms
ggml graph, CPUdefault elsewhere; CRISPASR_TITANET_GGML=1277.3 ms
ggml graph, GPUCRISPASR_TITANET_GGML=1 CRISPASR_TITANET_GPU=131.9 ms (but see below)

All three agree to cosine 1.000000 with each other and 0.999996 against NVIDIA's nemo_en_titanet_large.onnx export fed the same mel — so the choice is purely about speed.

⚠ CRISPASR_TITANET_GPU=1 is opt-in because it loses on real workloads despite winning the micro-benchmark. Diarization embeds one segment per call at variable lengths, so every call reshapes the graph and the GPU allocator re-reserves; and CRISPASR_SPEAKER_EMBED_WORKERS runs several embedders at once, which contend for the one GPU. End-to-end on a 600 s clip, 47 segments:

workers=4, legacy   9994 ms    <- default, fastest
workers=1, legacy  12673 ms
workers=1, GPU     15275 ms
workers=4, GPU     49866 ms

Keep it for evaluating a discrete GPU (where the balance may differ) or for CRISPASR_SPEAKER_EMBED_WORKERS=1 on a machine with weak CPU cores. Bucketing segment lengths so the graph shape stops changing is the work that would make this path win generally.

CRISPASR_TITANET_DUMP_MEL=<path> writes the computed mel as [T][n_mels] float32 — the counterpart to CRISPASR_TITANET_REF_MEL. Feeding that dump to an upstream ONNX export separates the front-end from the network, which a single end-to-end cosine cannot do.

VibeVoice

  • CRISPASR_VIBEVOICE_BENCH
  • CRISPASR_VIBEVOICE_DEBUG
  • CRISPASR_VIBEVOICE_DUMP_DIR
  • CRISPASR_VIBEVOICE_ENCODER_CHUNK_SECONDS
  • CRISPASR_VIBEVOICE_ENCODER_CONTEXT_SECONDS
  • CRISPASR_VIBEVOICE_LM_BUCKETS
  • CRISPASR_VIBEVOICE_NO_LM_BUCKETS
  • CRISPASR_VIBEVOICE_PRED_SCHED
  • CRISPASR_VIBEVOICE_QUANT_ALL
  • CRISPASR_VIBEVOICE_REF_FEATURES
  • CRISPASR_VIBEVOICE_REUSE_PRED_GRAPH
  • CRISPASR_VIBEVOICE_TTS_CFG_SCALE
  • CRISPASR_VIBEVOICE_TTS_DUMP
  • CRISPASR_VIBEVOICE_TTS_DUMP_DECODER
  • CRISPASR_VIBEVOICE_TTS_DUMP_PERFRAME
  • CRISPASR_VIBEVOICE_TTS_FLASH_ATTN
  • CRISPASR_VIBEVOICE_TTS_LATENTS
  • CRISPASR_VIBEVOICE_TTS_NOISE
  • CRISPASR_VIBEVOICE_TTS_SEED
  • CRISPASR_VIBEVOICE_TTS_TRACE
  • CRISPASR_VIBEVOICE_TTS_TRACE_FRAME
  • CRISPASR_VIBEVOICE_VAE_BACKEND
  • CRISPASR_VIBEVOICE_VOICE_AUDIO

VoxCPM2

  • CRISPASR_VOXCPM2_BENCH
  • CRISPASR_VOXCPM2_CFG_INTERVAL
  • CRISPASR_VOXCPM2_CFG_INTERVAL_DEBUG
  • CRISPASR_VOXCPM2_CFG_VALUE
  • CRISPASR_VOXCPM2_CPU_ONLY
  • CRISPASR_VOXCPM2_INFERENCE_STEPS
  • CRISPASR_VOXCPM2_MAX_LEN
  • CRISPASR_VOXCPM2_USE_REF
  • CRISPASR_VOXCPM2_VAE_MAX_SAMPLES - maximum 16 kHz input samples accepted by one voxcpm2-vae upscaling call (default 960000, or 60 seconds). Split longer audio, or raise this only when enough RAM/VRAM is available.

Voxtral / Voxtral-TTS

  • CRISPASR_VOXTRAL_BENCH
  • CRISPASR_VOXTRAL_FUSED_QKV
  • CRISPASR_VOXTRAL_TTS_CODEC_FROM_FILE
  • CRISPASR_VOXTRAL_TTS_DIFF_DUMP
  • CRISPASR_VOXTRAL_TTS_SEMANTIC_CB
  • CRISPASR_VOXTRAL_TTS_TEXT
  • CRISPASR_VOXTRAL_TTS_VOICE

Voxtral-4B

  • CRISPASR_VOXTRAL4B_BENCH
  • CRISPASR_VOXTRAL4B_FUSED_QKV
  • CRISPASR_VOXTRAL4B_STREAM_BATCH_ENCODER
  • CRISPASR_VOXTRAL4B_STREAM_CHUNK_MS
  • CRISPASR_VOXTRAL4B_STREAM_DEBUG
  • CRISPASR_VOXTRAL4B_STREAM_DECODER_THREAD
  • CRISPASR_VOXTRAL4B_STREAM_DIFF
  • CRISPASR_VOXTRAL4B_STREAM_LIVE
  • CRISPASR_VOXTRAL4B_STREAM_TIMING

Wav2Vec2

  • CRISPASR_WAV2VEC2_BENCH
  • CRISPASR_WAV2VEC2_DUMP_DIR
  • CRISPASR_WAV2VEC2_VERBOSE

WavTokenizer

  • CRISPASR_WAVTOK_BENCH
  • CRISPASR_WAVTOK_DUMP_DIR
  • CRISPASR_WAVTOK_FIXED_CODES

Zonos

  • CRISPASR_ZONOS_CPP_DUMP_DIR
  • CRISPASR_ZONOS_DECODE_CHUNK
  • CRISPASR_ZONOS_DECODE_CTX
  • CRISPASR_ZONOS_DIFF_N_STEPS
  • CRISPASR_ZONOS_FASTCONV
  • CRISPASR_ZONOS_SPEAKER_EMB_PATH
  • CRISPASR_ZONOS_TTS_BENCH
  • CRISPASR_ZONOS_TTS_TEXT