whisper.cpp

[!NOTE] Semantic versioning is still work in progress. More info can be found in https://github.com/ggml-org/ggml/discussions/1579

Nightly build: b4938

Change log since v1.9.2

371b5a75 release : v1.9.3 (#4000) 81c19056 cmake : update semver and release process [no ci] (#3996) 4834a232 talk-llama : sync llama.cpp 6b014cfb sync : ggml 8189458d ggml : bump version to 0.20.2 (ggml/1589) 51319a2c CUDA: MMVQ nwarps=8 for bs=1 for dense models on DGX Spark (llama/26843) 71759b7c cuda : skip UMA override for HIP builds (llama/27083) 964bb1b8 ggml : bump version to 0.20.1 (ggml/1587) 4257f470 sycl: fix thread/block count in quantized cpy kernel launches (llama/27160) 0f386138 support OP OPT_STEP_ADAMW, OPT_STEP_SGD (llama/25268) 9a0d1902 vulkan: add SHMEM_STRIDE_PAD/APPLY_SLM_A_RESHAPE for coopmat1 on Intel Xe (llama/25380) ab71410c fixed indent 9b98e57d Fixed gating logic for problematic Intel driver version 1fe009ca talk-llama : fix build (#0) 733f2819 sync : ggml 667da04a ggml : bump version to 0.20.0 (ggml/1584) b7ea8b19 ggml : recurrent state rollback for ggml_ssm_scan (llama/26623) 2aef2a00 sycl: fuse mul_mat(gate) + mul_mat(up) + GLU for q4_K dense FFN (llama/26779) 5bc0c852 ggml: force single thread on wasi (llama/25686) 2d05b6ec sycl: fuse the gated-delta-net state writeback cpy (llama/26643) 43cbe415 OpenVINO: Qwen3.5, memory optimization, and test-recurrent-state-rollback (llama/26952) 94d2d35a Support host pinned mem to improve SYCL Host-to-Device Memory Access (llama/26789) 34009e85 metal : add TQ2_0 support (llama/26980) 62031fed ggml-cpu/ops: vectorize flash-attention V-cache F16 to F32 conversion (llama/26947) ac9a74f1 sycl: remove separate fp32 type promotion in gemm non-oneDNN path (llama/26372) 3425d130 sycl: fuse UNARY(silu|sigmoid|softplus) + MUL (llama/26411) 90d4ed18 sycl : Add DMMV ESIMD Q3_K kernel (llama/26251) f7011606 sycl : enhance concat to support Q4_0, Q4_1, Q5_0, Q5_1, Q8_0 (llama/26800) c7be8b33 ggml-hip : remove -funsafe-math-optimizations (llama/26696) 406b1168 ggml : fix arm builds, unused var (llama/26991) b0e32971 gguf : harden loader against malformed tensor dims and metadata types (llama/25596) 92423afd kleidiai: Add runtime feature detection mechanism for aarch64/kleidiai (llama/26076) 1b93067b opencl: default FA c8 cluster width to 16 on X1E (llama/26433) 63290400 vulkan: add TQ2_0 (ternary) support (llama/25850) db3687b2 opencl: use flat mv q5_k when weight exceeds image1d_buffer_t limit (llama/26880) cdbe4551 CUDA: only disable CUDA graphs when mul_mat_id actually needs a stream sync (llama/26802) 030656f2 cuda : add warp-per-row wkv7 kernel for single-token decode (llama/26111) d9589688 llama: add default load-mode auto, which avoids mmap on iGPUs (llama/26081) 0f60f8e6 ggml-webgpu: fix CI errors from #25025 and #25262 (llama/26566) a6e2630f opencl: transpose the K tile in local memory for FA prefill kernels (llama/26428) 830ec225 ggml-cpu : fix CPU affinity mask being ignored on Android (llama/26838) b3bc9046 ggml : require contiguous src for ROLL on CUDA and Metal (llama/25928) f09a97cf ggml-webgpu : refactor several wgsl files and simplify flash_attn wgsl. (llama/26134) 877761c5 ggml-cpu : fix missing Q5_0 dispatch in SpaceMiT backend (llama/26792) 10791af6 CUDA: fuse rms_norm + mul + rope (+ view + set_rows) (llama/26767) 068d3b0b CUDA: fix thread/block count in quantized cpy kernel launches (llama/26731) eb3296f2 sycl: coalesce the ssm_conv window loads (llama/26612) 8a5ba01c metal : fix NORM/RMS_NORM for row lengths that leave a partial simdgroup (llama/26708) 06248ca1 cmake : add config version support (ggml/1582) 592feef0 talk-llama : sync llama.cpp 87704926 sync : ggml 84cdcad3 ggml : bump version to 0.19.0 (ggml/1581) 8587ad3b ggml : add aarch64 HWCAP fallbacks and fix fp16 variant detection (llama/25554) 56cb1547 sycl: fix UE4M3 parsing (llama/25608) 077c5d42 sycl: *glu flat path (llama/26354) 9faa9ee7 sycl : Support DSv4 OPs: LIGHTNING_INDEXER,DSV4_HC_COMB,DSV4_HC_POST,DSV4_HC_PRE (llama/26568) fb9e8ca9 sycl : fix error Error OP FLASH_ATTN_EXT on arc770 (llama/26441) 89d45afe sycl : enhance OP set_rows to support all missed data types (llama/26515) 79ab70c9 cuda: fix warnings for unused variable/function (llama/26688) 56933786 metal : avoid threadgroup matrix array instantiation in kernel_lightning_indexer (llama/26646) 69bd0a9a ci : onboard AMD ROCm CI with gfx1151 fixes (llama/26544) 5a80d0ad vulkan: fix submission batching size, add debug tools for diagnosing causes of DeviceLost drivers errors (llama/26371) 60894f1e mtmd/ggml: add ggml_build_forward_order (llama/26649) 87310219 vulkan backend ops: implemented GATED_LINEAR_ATTN (llama/25601) 8631825d whisper : heap out-of-bounds read in log_mel_spectrogram on very short audio (#3956) df1547b6 whisper,parakeet : reject invalid n_dims in tensor header to prevent stack-buffer-overflow on malformed model files (#3957)

What's Changed

• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3896
• include parakeet in build-xcframework.sh by @daaku in https://github.com/ggml-org/whisper.cpp/pull/3899
• Fix pkgconfig configuration (Nix build failure) by @nmouha in https://github.com/ggml-org/whisper.cpp/pull/3894
• examples : update model names in parakeet-cli README.md [no ci] by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3906
• examples: Fix argument flag for min speech duration in VAD by @QuantiusBenignus in https://github.com/ggml-org/whisper.cpp/pull/3907
• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3912
• examples : Improved inference performance of Android example project by @yymin1022 in https://github.com/ggml-org/whisper.cpp/pull/3913
• ci : set GGML_NATIVE=OFF for sanitize builds by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3920
• whisper : make voice_length() utf-8 aware for CJK by @buxuku in https://github.com/ggml-org/whisper.cpp/pull/3915
• whisper : map token timestamps to original time when VAD is enabled by @buxuku in https://github.com/ggml-org/whisper.cpp/pull/3910
• ci, examples : add package-lock.json to examples addon node by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3919
• whisper : expose internal VAD speech segments by @buxuku in https://github.com/ggml-org/whisper.cpp/pull/3916
• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3940
• Fix #587: Remove leading space from txt output by @JingliangGao in https://github.com/ggml-org/whisper.cpp/pull/3921
• ruby : add VAD speech segments API by @KitaitiMakoto in https://github.com/ggml-org/whisper.cpp/pull/3931
• Adding ruby bindings change log by @taf2 in https://github.com/ggml-org/whisper.cpp/pull/3959
• verify hparams loaded from parakeet model bin file by @bhargav191098 in https://github.com/ggml-org/whisper.cpp/pull/3950
• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3962
• Fix VAD min silence argument parsing by @chupakobra6 in https://github.com/ggml-org/whisper.cpp/pull/3963
• ci : set GGML_NATIVE=OFF for build-vad workflow by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3966
• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3969
• release : v1.9.2 by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3970

New Contributors

• @daaku made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3899
• @nmouha made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3894
• @QuantiusBenignus made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3907
• @yymin1022 made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3913
• @JingliangGao made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3921
• @bhargav191098 made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3950
• @chupakobra6 made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3963

Full Changelog: https://github.com/ggml-org/whisper.cpp/compare/v1.9.1...v1.9.2

What's Changed

• ci : add GGML_NATIVE=OFF and GGML_BMI2=OFF to windows-blas by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3891
• release : v1.9.1 by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3892

Full Changelog: https://github.com/ggml-org/whisper.cpp/compare/v1.9.0...v1.9.1

What's Changed

• ci : only trigger release jobs for tags by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3883
• parakeet : add support for NVIDIA Parakeet by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3735
• ruby : add support for Parakeet by @KitaitiMakoto in https://github.com/ggml-org/whisper.cpp/pull/3885
• release : v1.9.0 by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3886

Full Changelog: https://github.com/ggml-org/whisper.cpp/compare/v1.8.7...v1.9.0

What's Changed

• Do not assume /usr/lib library installation. by @waptaff in https://github.com/ggml-org/whisper.cpp/pull/3693
• server : merge split UTF-8 token text in verbose JSON by @lyonsno in https://github.com/ggml-org/whisper.cpp/pull/3850
• whisper : catch C++ exceptions in whisper_init_with_params_no_state by @danscMax in https://github.com/ggml-org/whisper.cpp/pull/3831
• ci : refactor + optimize by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3847
• ci : only publish/push docker images daily by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3854
• ci : use ccache instead of sccache for windows-cublas [no ci] by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3855
• ci : pin github actions to commit SHAs by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3856
• ci : use emscripten-core and pin version by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3857
• ci: build-windows action slimming by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3858
• Ci build quantize vad wasm ccache by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3860
• ci : add HF_TOKEN to docker.yml workflow [no ci] by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3861
• ci : add ccache to build-sycl [no ci] by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3859
• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3867
• coreml : fix --quantize crash and --optimize-ane label for mlprogram format by @krystophny in https://github.com/ggml-org/whisper.cpp/pull/3868
• ci : pin github actions to commit sha's by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3865
• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3880
• cli : add --version flag by @rumitvn in https://github.com/ggml-org/whisper.cpp/pull/3878
• release : v1.8.7 by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3881

New Contributors

• @waptaff made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3693
• @lyonsno made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3850
• @danscMax made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3831
• @krystophny made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3868
• @rumitvn made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3878

Full Changelog: https://github.com/ggml-org/whisper.cpp/compare/v1.8.6...v1.8.7

What's Changed

• common : re-implement ffmpeg-transcode.cpp + clarify ffmpeg usage by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3846
• ci : fix path to whisper.h in examples.yml [no ci] by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3842

Full Changelog: https://github.com/ggml-org/whisper.cpp/compare/v1.8.5...v1.8.6

Overview

Maintenance release + performance improvements all around:

https://github.com/ggml-org/whisper.cpp/blob/master/scripts/bench-all-gg.txt

What's Changed

• ruby : fix dangling pointers, memory leak, and SEGV on parallel transcription by @KitaitiMakoto in https://github.com/ggml-org/whisper.cpp/pull/3715
• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3730
• vad : add streaming detect + explicit state reset by @danielbodart in https://github.com/ggml-org/whisper.cpp/pull/3677
• bench : sync submit-results URL to ggml-org by @jinweihan-ai in https://github.com/ggml-org/whisper.cpp/pull/3769
• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3786
• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3789
• ruby : transcribe without GVL, accept more MemoryViews, Windows support, fix memory size report, improve document by @KitaitiMakoto in https://github.com/ggml-org/whisper.cpp/pull/3775
• Incorrect timestamps by @bviksoe in https://github.com/ggml-org/whisper.cpp/pull/2279
• server: Add support for controlling token_timestamps directly by @alubbe in https://github.com/ggml-org/whisper.cpp/pull/3785
• Fix #984: About max_tokens skipping remaining audio issue by @annaeina in https://github.com/ggml-org/whisper.cpp/pull/3798
• server: Fix params leak between requests by @alubbe in https://github.com/ggml-org/whisper.cpp/pull/3784
• server: Fix no_speech_thold not being read by @alubbe in https://github.com/ggml-org/whisper.cpp/pull/3783
• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3803
• server: Add support for carry_initial_prompt by @alubbe in https://github.com/ggml-org/whisper.cpp/pull/3781
• server: Return speaker information in JSON by @alubbe in https://github.com/ggml-org/whisper.cpp/pull/3782
• examples : fix memory leak in read_audio_data by @petterreinholdtsen in https://github.com/ggml-org/whisper.cpp/pull/3810
• whisper : set bench data for each iteration by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3812
• ci : use github ubuntu-22.04-arm runner instead of qemu by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3815
• fix: server /inference fails to decode in-memory audio (regression) by @ServeurpersoCom in https://github.com/ggml-org/whisper.cpp/pull/3818
• fix: in bindings/ruby/test/jfk_reader/jfk_reader in jfk_reader.c by @orbisai0security in https://github.com/ggml-org/whisper.cpp/pull/3756
• cmake : add CMakePresets.json [no ci] by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3808
• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3824
• [AMD] Add AMD ROCm/HIP GPU build instructions by @Kaihui-AMD in https://github.com/ggml-org/whisper.cpp/pull/3823
• cli : merge tokens split across UTF-8 boundaries in JSON output by @texasich in https://github.com/ggml-org/whisper.cpp/pull/3751
• docs : add AGENTS.md and CONTRIBUTING.md [no ci] by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3826
• ci : only run docker jobs when pushed to master [no ci] by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3828
• ci : set GGML_NATIVE=OFF for bindings-java by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3830
• ci : renable arm64 docker builds by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3832
• ci : add on push/pull_request paths ruby job by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3833
• ci : fix include paths for bindings-go job [no ci] by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3835
• ci : add ignore for bindings/{ruby, go} in build.yml [no ci] by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3837
• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3840

New Contributors

• @danielbodart made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3677
• @jinweihan-ai made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3769
• @bviksoe made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/2279
• @annaeina made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3798
• @ServeurpersoCom made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3818
• @orbisai0security made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3756
• @Kaihui-AMD made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3823
• @texasich made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3751

Full Changelog: https://github.com/ggml-org/whisper.cpp/compare/v1.8.4...v1.8.5

Overview

Maintenance release, latest ggml, some performance gains across the board.

What's Changed

• Fix: UTF-8 character cut off to two "�" in segment wrapping (max_len) by @lordofriver in https://github.com/ggml-org/whisper.cpp/pull/3592
• examples: add -g/--gpu-device option and GPU_DEVICE env support by @brauliobo in https://github.com/ggml-org/whisper.cpp/pull/3557
• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3636
• Fix macOS dSYM path case mismatch in XCFramework build by @friederbluemle in https://github.com/ggml-org/whisper.cpp/pull/3630
• ruby : add VAD::Context#segments_from_samples, allow Pathname, etc. by @KitaitiMakoto in https://github.com/ggml-org/whisper.cpp/pull/3633
• ruby : add Whisper::Context::Params, fix token memory management by @KitaitiMakoto in https://github.com/ggml-org/whisper.cpp/pull/3647
• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3652
• ci : try fix mirrors by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3655
• server : fix hardcoded /inference path in default HTML page by @sidmohan0 in https://github.com/ggml-org/whisper.cpp/pull/3639
• cmake: Drop obsolete build-time configuration of backends by @ckastner in https://github.com/ggml-org/whisper.cpp/pull/3649
• chore: Update outdated GitHub Actions versions by @pgoslatara in https://github.com/ggml-org/whisper.cpp/pull/3646
• ci: add vulkan docker image by @rare-magma in https://github.com/ggml-org/whisper.cpp/pull/3644
• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3665
• readme : fix duplicate word typo in VAD section by @cluster2600 in https://github.com/ggml-org/whisper.cpp/pull/3670
• cli: update miniaudio library to 0.11.24 by @data-man in https://github.com/ggml-org/whisper.cpp/pull/3672
• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3684
• ruby : null-check by @KitaitiMakoto in https://github.com/ggml-org/whisper.cpp/pull/3689
• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3710
• server : return proper HTTP status codes for error responses by @dearlordylord in https://github.com/ggml-org/whisper.cpp/pull/3707
• server: allow max_len to be used for any output format by @gaelj in https://github.com/ggml-org/whisper.cpp/pull/3679
• fix: replace deprecated openvino-dev with openvino>=2023.3.0 for Python 3.12+ compat by @Aiudadadadf in https://github.com/ggml-org/whisper.cpp/pull/3678
• fix(go): handle EOF correctly in model download example of golang bidings by @Lumberj3ck in https://github.com/ggml-org/whisper.cpp/pull/3671
• fix: VAD time mapping timestamp drift caused by overlap samples by @lohopupa in https://github.com/ggml-org/whisper.cpp/pull/3711
• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3713

New Contributors

• @lordofriver made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3592
• @brauliobo made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3557
• @friederbluemle made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3630
• @sidmohan0 made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3639
• @ckastner made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3649
• @pgoslatara made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3646
• @rare-magma made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3644
• @cluster2600 made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3670
• @dearlordylord made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3707
• @gaelj made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3679
• @Aiudadadadf made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3678
• @Lumberj3ck made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3671
• @lohopupa made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3711

Full Changelog: https://github.com/ggml-org/whisper.cpp/compare/v1.8.3...v1.8.4

Overview

Maintenance release, latest ggml, minor improvements in the tools/server/bindings.

What's Changed

• server : set no_context == true by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3482
• WASM example broken for Hebrew by @Orel-A in https://github.com/ggml-org/whisper.cpp/pull/3487
• ruby : tiny bug fix by @KitaitiMakoto in https://github.com/ggml-org/whisper.cpp/pull/3490
• 12x Performance Boost with Integrated Graphics by @OrelSokolov in https://github.com/ggml-org/whisper.cpp/pull/3492
• ruby : VAD separately from ASR by @KitaitiMakoto in https://github.com/ggml-org/whisper.cpp/pull/3518
• vad : Silero VAD v6.2.0 by @KitaitiMakoto in https://github.com/ggml-org/whisper.cpp/pull/3524
• need to #include to compile on macos 11 by @smilingpoplar in https://github.com/ggml-org/whisper.cpp/pull/3533
• Add verified device of Ascend NPU by @lrisguan in https://github.com/ggml-org/whisper.cpp/pull/3516
• examples : fix typo in vad-speech-segments command [no ci] by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3535
• tests : update VAD tests to use Silero V6.2.0 by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3534
• vad : fix buffer overflow in sample reduction loop by @josephsellers in https://github.com/ggml-org/whisper.cpp/pull/3558
• Allow NextSegment() to be called across multiple Process() calls in the Go binding by @joshmux in https://github.com/ggml-org/whisper.cpp/pull/3503
• Enable VAD for Go bindings by @joshmux in https://github.com/ggml-org/whisper.cpp/pull/3563
• server: allow custom temp directory for ffmpeg by @socram8888 in https://github.com/ggml-org/whisper.cpp/pull/3564
• Fix FreeBSD linker error: undefined reference to pthread_create by @rsmirnov90 in https://github.com/ggml-org/whisper.cpp/pull/3568
• ruby : add Whisper::Token, fix model URI by @KitaitiMakoto in https://github.com/ggml-org/whisper.cpp/pull/3575
• ruby : fix segmentation fault by @KitaitiMakoto in https://github.com/ggml-org/whisper.cpp/pull/3591
• cmake: fix executable example targets by @ink-splatters in https://github.com/ggml-org/whisper.cpp/pull/3600

New Contributors

• @Orel-A made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3487
• @OrelSokolov made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3492
• @smilingpoplar made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3533
• @lrisguan made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3516
• @josephsellers made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3558
• @joshmux made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3503
• @socram8888 made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3564
• @rsmirnov90 made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3568
• @ink-splatters made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3600

Full Changelog: https://github.com/ggml-org/whisper.cpp/compare/v1.8.2...v1.8.3

Overview

• Fix a bug in the ggml norm CPU scalar operator

What's Changed

• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3478

Full Changelog: https://github.com/ggml-org/whisper.cpp/compare/v1.8.1...v1.8.2

Overview

• Fix Vulkan builds
• Fix memory leaks when using VAD
• Support --carry-initial-prompt

What's Changed

• bindings-java : disable flash attention by default by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3445
• ruby : Loose RegExp of a test to follow change of flash attension default by @KitaitiMakoto in https://github.com/ggml-org/whisper.cpp/pull/3448
• vad : fix memory leaks in VAD implementation by @danbev in https://github.com/ggml-org/whisper.cpp/pull/3453
• ruby : Bump Ruby bindings' version to 1.3.4 by @KitaitiMakoto in https://github.com/ggml-org/whisper.cpp/pull/3461
• Fix vad leaks by @silviucpp in https://github.com/ggml-org/whisper.cpp/pull/3463
• cli: Fix assignment for vad_min_silence_duration_ms by @alubbe in https://github.com/ggml-org/whisper.cpp/pull/3467
• Add support for --carry-initial-prompt by @alubbe in https://github.com/ggml-org/whisper.cpp/pull/3395
• Support using devices of type iGPU by @0cc4m in https://github.com/ggml-org/whisper.cpp/pull/3469
• sync : ggml by @ggerganov in https://github.com/ggml-org/whisper.cpp/pull/3470

New Contributors

• @silviucpp made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3463
• @0cc4m made their first contribution in https://github.com/ggml-org/whisper.cpp/pull/3469

Full Changelog: https://github.com/ggml-org/whisper.cpp/compare/v1.8.0...v1.8.1