source languages: nl; target languages: en; OPUS readme: nl-en; dataset: opus; model: transformer-align; pre-processing: normalization + SentencePiece.
Open-weight model · Translation
Index-Echo-S2ST-9B-FP8
by Index Team IndexTeam/Index-Echo-S2ST-9B-FP8
Index-Echo-S2ST-9B-FP8 is an open-weight model for translation from Index Team, released under Apache License 2.0. Its published files total 527.3 KB.
Official FP8 (W8A8) quantization of IndexTeam/Index-Echo-S2ST-9B, part of the Index-Echo speech-to-speech translation (S2ST) model family by bilibili.
Parameters—
Context—
Weights527.3 KB
Licenseapache-2.0
AccessOpen weights
Monthly Downloads—
Model Card
By Index Team, published under apache-2.0, revision 735ccd0ce1c7.
Official FP8 (W8A8) quantization of IndexTeam/Index-Echo-S2ST-9B, part of the Index-Echo speech-to-speech translation (S2ST) model family by bilibili. This repository mirrors the original checkpoint layout; only the text LLM backbone (stlmllm/) is quantized to FP8 - the audio tower, connector, and all speech-synthesis components remain in BF16. Use it exactly like the original repo (same infer.py / configs). - All Linear layers of the language model backbone (stlmllm/) are quantized; the audio tower, connector, lmhead, embeddings and all other pipeline components are kept in BF16. Measured on an NVIDIA A100 against the original BF16 checkpoint (greedy decoding, official translation prompt)…
Read Index Team's full model card
Official FP8 (W8A8) quantization of IndexTeam/Index-Echo-S2ST-9B, part of the Index-Echo speech-to-speech translation (S2ST) model family by bilibili.
This repository mirrors the original checkpoint layout; only the text LLM backbone (stlm_llm/) is quantized to FP8 - the audio tower, connector, and all speech-synthesis components remain in BF16. Use it exactly like the original repo (same infer.py / configs).
Quantization
- Scheme:
FP8_DYNAMIC(FP8 E4M3 weights, per-token dynamic FP8 activations), produced with llm-compressor (quantization_schemerecorded inrecipe.yaml). - All
Linearlayers of the language model backbone (stlm_llm/) are quantized; the audio tower, connector,lm_head, embeddings and all other pipeline components are kept in BF16. - Format: compressed-tensors safetensors - load directly with vLLM (
quantization="compressed-tensors") or transformers.
Consistency validation
Measured on an NVIDIA A100 against the original BF16 checkpoint (greedy decoding, official translation prompt):
| Metric | BF16 | FP8 | Delta |
|---|---|---|---|
| Perplexity (fixed corpus) | 3.8218 | 3.8412 | +0.51% |
| zh->en generation identical | - | - | no (semantically equivalent) |
| en->zh generation identical | - | - | no (semantically equivalent) |
Usage
Identical to the original checkpoint - clone this repo and follow the README / infer.py of the base model (IndexTeam/Index-Echo-S2ST-9B). The quantized LLM backbone loads via compressed-tensors; make sure compressed-tensors (or vLLM / a recent transformers) is installed.
Quantized and published by the Index team, 2026-10-03.
Configuration
- Model type
- dubbing_bridge
Identity and Version
- Repository
- IndexTeam/Index-Echo-S2ST-9B-FP8
- Publisher
- Index Team
- Task
- Translation
- Modality
- Text
- Library
- transformers
- Parameters
- Not stated by the source
- Languages
- Not stated by the source
- Revision
- 735ccd0ce1c7c0bb85bdc6fb4839b1c251f12547
- First published
- 2026-10-02
- Last updated
- 2026-10-03
Files and Weights
256 files, 527.3 KB in total.
Configuration174 files · 366.0 KB
Tokenizer2 files · 2.4 KB
Documentation15 files · 34.8 KB
Other56 files · 117.7 KB
Repository9 files · 6.4 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| bridge/mapper_config.json | Configuration | 171 B | — |
| code/bridge/text_norm.py | Configuration | 4.0 KB | — |
| code/cosyvoice_repo/.github/workflows/lint.yml | Configuration | 2.3 KB | — |
| code/cosyvoice_repo/.github/workflows/stale-issues.yml | Configuration | 705 B | — |
| code/cosyvoice_repo/cosyvoice/__init__.py | Configuration | — | |
| code/cosyvoice_repo/cosyvoice/bin/average_model.py | Configuration | 3.2 KB | — |
| code/cosyvoice_repo/cosyvoice/bin/export_jit.py | Configuration | 3.8 KB | — |
| code/cosyvoice_repo/cosyvoice/bin/export_onnx.py | Configuration | 4.5 KB | — |
| code/cosyvoice_repo/cosyvoice/cli/__init__.py | Configuration | — | |
| code/cosyvoice_repo/cosyvoice/dataset/__init__.py | Configuration | — | |
| code/cosyvoice_repo/cosyvoice/dataset/dataset.py | Configuration | 5.1 KB | — |
| code/cosyvoice_repo/cosyvoice/flow/DiT/dit.py | Configuration | 5.7 KB | — |
| code/cosyvoice_repo/cosyvoice/flow/length_regulator.py | Configuration | 3.1 KB | — |
| code/cosyvoice_repo/cosyvoice/hifigan/f0_predictor.py | Configuration | 3.8 KB | — |
| code/cosyvoice_repo/cosyvoice/hifigan/hifigan.py | Configuration | 3.2 KB | — |
| code/cosyvoice_repo/cosyvoice/transformer/__init__.py | Configuration | — | |
| code/cosyvoice_repo/cosyvoice/transformer/activation.py | Configuration | 3.1 KB | — |
| code/cosyvoice_repo/cosyvoice/transformer/decoder_layer.py | Configuration | 4.8 KB | — |
| code/cosyvoice_repo/cosyvoice/transformer/label_smoothing_loss.py | Configuration | 3.5 KB | — |
| code/cosyvoice_repo/cosyvoice/transformer/positionwise_feed_forward.py | Configuration | 4.2 KB | — |
| code/cosyvoice_repo/cosyvoice/utils/__init__.py | Configuration | — | |
| code/cosyvoice_repo/cosyvoice/utils/class_utils.py | Configuration | 3.6 KB | — |
| code/cosyvoice_repo/cosyvoice/utils/file_utils.py | Configuration | 5.1 KB | — |
| code/cosyvoice_repo/cosyvoice/utils/frontend_utils.py | Configuration | 4.3 KB | — |
| code/cosyvoice_repo/cosyvoice/utils/losses.py | Configuration | 2.1 KB | — |
| code/cosyvoice_repo/cosyvoice/utils/onnx.py | Configuration | 2.9 KB | — |
| code/cosyvoice_repo/cosyvoice/vllm/cosyvoice2.py | Configuration | 4.5 KB | — |
| code/cosyvoice_repo/examples/grpo/cosyvoice2/huggingface_to_pretrained.py | Configuration | 2.9 KB | — |
| code/cosyvoice_repo/examples/grpo/cosyvoice2/prepare_data.py | Configuration | 3.0 KB | — |
| code/cosyvoice_repo/examples/grpo/cosyvoice2/pretrained_to_huggingface.py | Configuration | 5.5 KB | — |
| code/cosyvoice_repo/examples/libritts/cosyvoice/conf/ds_stage2.json | Configuration | 925 B | — |
| code/cosyvoice_repo/examples/libritts/cosyvoice/local/prepare_data.py | Configuration | 1.9 KB | — |
| code/cosyvoice_repo/examples/libritts/cosyvoice/local/prepare_reject_sample.py | Configuration | 1.8 KB | — |
| code/cosyvoice_repo/examples/libritts/cosyvoice/tts_text.json | Configuration | 89 B | — |
| code/cosyvoice_repo/examples/libritts/cosyvoice2/conf/ds_stage2.json | Configuration | 925 B | — |
| code/cosyvoice_repo/examples/libritts/cosyvoice2/tts_text.json | Configuration | 89 B | — |
| code/cosyvoice_repo/examples/libritts/cosyvoice3/conf/cosyvoice3.yaml | Configuration | 7.1 KB | — |
| code/cosyvoice_repo/examples/libritts/cosyvoice3/conf/ds_stage2.json | Configuration | 925 B | — |
| code/cosyvoice_repo/examples/magicdata-read/cosyvoice/local/prepare_data.py | Configuration | 1.7 KB | — |
| code/cosyvoice_repo/examples/magicdata-read/cosyvoice/tts_text.json | Configuration | 598 B | — |
| code/cosyvoice_repo/runtime/python/fastapi/client.py | Configuration | 3.8 KB | — |
| code/cosyvoice_repo/runtime/python/fastapi/server.py | Configuration | 3.7 KB | — |
| code/cosyvoice_repo/runtime/python/grpc/client.py | Configuration | 4.8 KB | — |
| code/cosyvoice_repo/runtime/python/grpc/server.py | Configuration | 4.3 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/client_http.py | Configuration | 5.6 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/docker-compose.cosyvoice2.dit.yml | Configuration | 735 B | — |
| code/cosyvoice_repo/runtime/triton_trtllm/docker-compose.cosyvoice2.unet.yml | Configuration | 630 B | — |
| code/cosyvoice_repo/runtime/triton_trtllm/docker-compose.cosyvoice3.yml | Configuration | 615 B | — |
| code/cosyvoice_repo/runtime/triton_trtllm/model_repo/audio_tokenizer/1/model.py | Configuration | 4.0 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/model_repo/speaker_embedding/1/model.py | Configuration | 6.8 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/model_repo/token2wav_dit/1/model.py | Configuration | 5.5 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/model_repo_cosyvoice3/audio_tokenizer/1/model.py | Configuration | 3.8 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/model_repo_cosyvoice3/speaker_embedding/1/model.py | Configuration | 6.6 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/model_repo_cosyvoice3/vocoder/1/model.py | Configuration | 2.6 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/scripts/fill_template.py | Configuration | 1.8 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/scripts/test_llm.py | Configuration | 5.1 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/streaming_inference.py | Configuration | 5.3 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/.github/codecov.yml | Configuration | 377 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/.github/dependabot.yml | Configuration | 731 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/.github/release-drafter.yml | Configuration | 784 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/.pre-commit-config.yaml | Configuration | 1.5 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/__init__.py | Configuration | 81 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/callbacks/default.yaml | Configuration | 97 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/callbacks/model_checkpoint.yaml | Configuration | 1.2 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/callbacks/model_summary.yaml | Configuration | 252 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/callbacks/none.yaml | Configuration | — | |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/callbacks/rich_progress_bar.yaml | Configuration | 172 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/data/hi-fi_en-US_female.yaml | Configuration | 472 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/data/ljspeech.yaml | Configuration | 520 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/data/vctk.yaml | Configuration | 385 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/debug/default.yaml | Configuration | 903 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/debug/fdr.yaml | Configuration | 120 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/debug/limit.yaml | Configuration | 218 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/debug/overfit.yaml | Configuration | 204 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/debug/profiler.yaml | Configuration | 225 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/eval.yaml | Configuration | 335 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/experiment/hifi_dataset_piper_phonemizer.yaml | Configuration | 423 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/experiment/ljspeech.yaml | Configuration | 332 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/experiment/ljspeech_min_memory.yaml | Configuration | 361 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/experiment/multispeaker.yaml | Configuration | 336 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/extras/default.yaml | Configuration | 232 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/hparams_search/mnist_optuna.yaml | Configuration | 1.8 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/hydra/default.yaml | Configuration | 608 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/logger/aim.yaml | Configuration | 1.3 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/logger/comet.yaml | Configuration | 372 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/logger/csv.yaml | Configuration | 157 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/logger/many_loggers.yaml | Configuration | 118 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/logger/mlflow.yaml | Configuration | 339 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/logger/neptune.yaml | Configuration | 277 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/logger/tensorboard.yaml | Configuration | 258 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/logger/wandb.yaml | Configuration | 522 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/model/cfm/default.yaml | Configuration | 40 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/model/decoder/default.yaml | Configuration | 119 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/model/encoder/default.yaml | Configuration | 417 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/model/matcha.yaml | Configuration | 328 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/model/optimizer/adam.yaml | Configuration | 70 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/paths/default.yaml | Configuration | 632 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/train.yaml | Configuration | 1.6 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/trainer/cpu.yaml | Configuration | 51 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/trainer/ddp.yaml | Configuration | 104 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/trainer/ddp_sim.yaml | Configuration | 115 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/trainer/default.yaml | Configuration | 439 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/trainer/gpu.yaml | Configuration | 51 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/trainer/mps.yaml | Configuration | 51 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/__init__.py | Configuration | — | |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/data/__init__.py | Configuration | — | |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/data/components/__init__.py | Configuration | — | |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/hifigan/__init__.py | Configuration | — | |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/hifigan/config.py | Configuration | 779 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/hifigan/denoiser.py | Configuration | 2.6 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/hifigan/env.py | Configuration | 429 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/hifigan/meldataset.py | Configuration | 6.8 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/hifigan/xutils.py | Configuration | 1.4 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/models/__init__.py | Configuration | — | |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/models/baselightningmodule.py | Configuration | 7.0 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/models/components/__init__.py | Configuration | — | |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/models/components/flow_matching.py | Configuration | 4.7 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/onnx/__init__.py | Configuration | — | |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/onnx/export.py | Configuration | 5.4 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/onnx/infer.py | Configuration | 6.3 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/text/__init__.py | Configuration | 1.7 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/text/cleaners.py | Configuration | 3.6 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/text/numbers.py | Configuration | 2.2 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/text/symbols.py | Configuration | 635 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/train.py | Configuration | 4.6 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/utils/__init__.py | Configuration | 326 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/utils/audio.py | Configuration | 2.3 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/utils/generate_data_statistics.py | Configuration | 3.3 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/utils/instantiators.py | Configuration | 1.8 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/utils/logging_utils.py | Configuration | 1.7 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/utils/model.py | Configuration | 2.9 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/utils/monotonic_align/__init__.py | Configuration | 646 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/utils/monotonic_align/setup.py | Configuration | 207 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/utils/pylogger.py | Configuration | 720 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/utils/rich_utils.py | Configuration | 3.3 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/setup.py | Configuration | 1.5 KB | — |
| code/cosyvoice_repo/tools/extract_embedding.py | Configuration | 3.0 KB | — |
| code/cosyvoice_repo/tools/extract_speech_token.py | Configuration | 2.8 KB | — |
| code/cosyvoice_repo/tools/make_parquet_list.py | Configuration | 5.7 KB | — |
| code/cosyvoice_repo/vllm_example.py | Configuration | 1.7 KB | — |
| code/ja_ext/deprecated/__init__.py | Configuration | 398 B | — |
| code/ja_ext/deprecated/params.py | Configuration | 2.9 KB | — |
| code/ja_ext/jaconv/__init__.py | Configuration | 2.7 KB | — |
| code/ja_ext/jaconv/compat.py | Configuration | 241 B | — |
| code/ja_ext/jaconv/conv_table.py | Configuration | 5.7 KB | — |
| code/ja_ext/pykakasi/__init__.py | Configuration | 1.9 KB | — |
| code/ja_ext/pykakasi/cli.py | Configuration | 4.5 KB | — |
| code/ja_ext/pykakasi/kakasi.py | Configuration | 6.1 KB | — |
| code/ja_ext/pykakasi/kanji.py | Configuration | 3.4 KB | — |
| code/ja_ext/pyopenjtalk/utils.py | Configuration | 835 B | — |
| code/ja_ext/pyopenjtalk/version.py | Configuration | 22 B | — |
| code/ja_ext/wrapt/__init__.py | Configuration | 2.9 KB | — |
| code/ja_ext/wrapt/__wrapt__.py | Configuration | 1.4 KB | — |
| code/ja_ext/wrapt/arguments.py | Configuration | 2.3 KB | — |
| code/ja_ext/wrapt/exceptions.py | Configuration | 2.3 KB | — |
| code/ja_ext/wrapt/weakrefs.py | Configuration | 5.2 KB | — |
| code/stlm/audio_model.py | Configuration | 5.8 KB | — |
| code/stlm/stlm_api.py | Configuration | 4.1 KB | — |
| code/tts/ja_kata.py | Configuration | 3.9 KB | — |
| config.json | Configuration | 4.2 KB | — |
| cosyvoice3/CosyVoice-BlankEN/config.json | Configuration | 659 B | — |
| cosyvoice3/CosyVoice-BlankEN/generation_config.json | Configuration | 242 B | — |
| cosyvoice3/config.json | Configuration | 2 B | — |
| cosyvoice3/configuration.json | Configuration | 47 B | — |
| cosyvoice3/cosyvoice3.yaml | Configuration | 6.9 KB | — |
| modeling_dubbing.py | Configuration | 3.0 KB | — |
| samples/refs.json | Configuration | 478 B | — |
| samples/verify_report.json | Configuration | 3.4 KB | — |
| stlm_llm/config.json | Configuration | 3.2 KB | — |
| stlm_llm/generation_config.json | Configuration | 116 B | — |
| stlm_llm/recipe.yaml | Configuration | 281 B | — |
| stlm_llm/validation.json | Configuration | 5.1 KB | — |
| validation.json | Configuration | 5.1 KB | — |
| verify_export.py | Configuration | 5.8 KB | — |
| README.md | Documentation | 2.0 KB | — |
| code/cosyvoice_repo/.github/ISSUE_TEMPLATE/bug_report.md | Documentation | 834 B | — |
| code/cosyvoice_repo/.github/ISSUE_TEMPLATE/feature_request.md | Documentation | 595 B | — |
| code/cosyvoice_repo/CODE_OF_CONDUCT.md | Documentation | 3.4 KB | — |
| code/cosyvoice_repo/FAQ.md | Documentation | 671 B | — |
| code/cosyvoice_repo/examples/grpo/cosyvoice2/README.md | Documentation | 4.6 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/README.Cosyvoice2.Unet.md | Documentation | 6.4 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/README.Cosyvoice3.md | Documentation | 3.4 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/README.md | Documentation | 1.5 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/.github/PULL_REQUEST_TEMPLATE.md | Documentation | 803 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/LICENSE | Documentation | 1.1 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/hifigan/LICENSE | Documentation | 1.1 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/hifigan/README.md | Documentation | 5.6 KB | — |
| code/ja_ext/pyopenjtalk/htsvoice/LICENSE_mei_normal.htsvoice | Documentation | 2.9 KB | — |
| code/ja_ext/pyopenjtalk/htsvoice/README.md | Documentation | 100 B | — |
| code/cosyvoice_repo/docker/Dockerfile | Other | 2.1 KB | — |
| code/cosyvoice_repo/examples/grpo/cosyvoice2/Dockerfile | Other | 486 B | — |
| code/cosyvoice_repo/examples/grpo/cosyvoice2/requirements.txt | Other | 457 B | — |
| code/cosyvoice_repo/examples/grpo/cosyvoice2/run.sh | Other | 6.2 KB | — |
| code/cosyvoice_repo/examples/grpo/cosyvoice2/scripts/compute_wer.sh | Other | 942 B | — |
| code/cosyvoice_repo/examples/libritts/cosyvoice/local/download_and_untar.sh | Other | 2.8 KB | — |
| code/cosyvoice_repo/examples/libritts/cosyvoice/path.sh | Other | 185 B | — |
| code/cosyvoice_repo/examples/libritts/cosyvoice/run.sh | Other | 4.4 KB | — |
| code/cosyvoice_repo/examples/libritts/cosyvoice2/path.sh | Other | 185 B | — |
| code/cosyvoice_repo/examples/libritts/cosyvoice2/run.sh | Other | 4.6 KB | — |
| code/cosyvoice_repo/examples/libritts/cosyvoice2/run_dpo.sh | Other | 5.1 KB | — |
| code/cosyvoice_repo/examples/libritts/cosyvoice3/path.sh | Other | 185 B | — |
| code/cosyvoice_repo/examples/libritts/cosyvoice3/run.sh | Other | 4.7 KB | — |
| code/cosyvoice_repo/examples/magicdata-read/cosyvoice/local/download_and_untar.sh | Other | 2.6 KB | — |
| code/cosyvoice_repo/examples/magicdata-read/cosyvoice/path.sh | Other | 185 B | — |
| code/cosyvoice_repo/examples/magicdata-read/cosyvoice/run.sh | Other | 4.0 KB | — |
| code/cosyvoice_repo/requirements.txt | Other | 1.1 KB | — |
| code/cosyvoice_repo/runtime/python/Dockerfile | Other | 732 B | — |
| code/cosyvoice_repo/runtime/python/grpc/cosyvoice.proto | Other | 757 B | — |
| code/cosyvoice_repo/runtime/triton_trtllm/Dockerfile.server | Other | 411 B | — |
| code/cosyvoice_repo/runtime/triton_trtllm/model_repo/audio_tokenizer/config.pbtxt | Other | 1.2 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/model_repo/cosyvoice2/config.pbtxt | Other | 1.5 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/model_repo/cosyvoice2_dit/config.pbtxt | Other | 1.5 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/model_repo/speaker_embedding/config.pbtxt | Other | 1.1 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/model_repo/token2wav/config.pbtxt | Other | 1.6 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/model_repo/token2wav_dit/config.pbtxt | Other | 1.4 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/model_repo_cosyvoice3/audio_tokenizer/config.pbtxt | Other | 1.2 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/model_repo_cosyvoice3/cosyvoice3/config.pbtxt | Other | 1.5 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/model_repo_cosyvoice3/speaker_embedding/config.pbtxt | Other | 1.1 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/model_repo_cosyvoice3/token2wav/config.pbtxt | Other | 1.1 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/model_repo_cosyvoice3/vocoder/config.pbtxt | Other | 567 B | — |
| code/cosyvoice_repo/runtime/triton_trtllm/requirements.txt | Other | 161 B | — |
| code/cosyvoice_repo/runtime/triton_trtllm/run.sh | Other | 6.8 KB | — |
| code/cosyvoice_repo/runtime/triton_trtllm/run_cosyvoice3.sh | Other | 6.6 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/MANIFEST.in | Other | 352 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/Makefile | Other | 1.2 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/VERSION | Other | 8 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/matcha/utils/monotonic_align/core.pyx | Other | 1.2 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/pyproject.toml | Other | 982 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/requirements.txt | Other | 904 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/scripts/schedule.sh | Other | 207 B | — |
| code/ja_ext/jaconv/__init__.pyi | Other | 955 B | — |
| code/ja_ext/jaconv/compat.pyi | Other | 231 B | — |
| code/ja_ext/jaconv/conv_table.pyi | Other | 533 B | — |
| code/ja_ext/jaconv/jaconv.pyi | Other | 2.7 KB | — |
| code/ja_ext/jaconv/py.typed | Other | — | |
| code/ja_ext/pykakasi/data/halfkana3.db | Other | 1.3 KB | — |
| code/ja_ext/pykakasi/data/hepburndict3.db | Other | 6.8 KB | eeeaebd8d1a7 |
| code/ja_ext/pykakasi/data/hepburnhira3.db | Other | 3.3 KB | 80375d807672 |
| code/ja_ext/pykakasi/data/kunreidict3.db | Other | 6.8 KB | bd77affee1d7 |
| code/ja_ext/pykakasi/data/kunreihira3.db | Other | 3.3 KB | 5ab1fdcf445c |
| code/ja_ext/pykakasi/data/passporthira3.db | Other | 3.7 KB | 9e8e8cdf0061 |
| code/ja_ext/pykakasi/py.typed | Other | 27 B | — |
| code/ja_ext/pyopenjtalk/open_jtalk_dic_utf_8-1.11/COPYING | Other | 5.9 KB | — |
| code/ja_ext/pyopenjtalk/open_jtalk_dic_utf_8-1.11/pos-id.def | Other | 1.9 KB | — |
| code/ja_ext/pyopenjtalk/open_jtalk_dic_utf_8-1.11/unk.dic | Other | 5.7 KB | — |
| .gitattributes | Repository | 3.0 KB | — |
| code/cosyvoice_repo/.gitignore | Repository | 560 B | — |
| code/cosyvoice_repo/.gitmodules | Repository | 123 B | — |
| code/cosyvoice_repo/runtime/triton_trtllm/model_repo/tensorrt_llm/1/.gitkeep | Repository | — | |
| code/cosyvoice_repo/third_party/Matcha-TTS/.env.example | Repository | 346 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/.gitignore | Repository | 2.2 KB | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/.project-root | Repository | 81 B | — |
| code/cosyvoice_repo/third_party/Matcha-TTS/configs/local/.gitkeep | Repository | — | |
| code/cosyvoice_repo/third_party/Matcha-TTS/notebooks/.gitkeep | Repository | — | |
| cosyvoice3/CosyVoice-BlankEN/tokenizer_config.json | Tokenizer | 1.3 KB | — |
| stlm_llm/tokenizer_config.json | Tokenizer | 1.1 KB | — |
License and Download
- License
- apache-2.0
- Access
- Open weights, no gate
Download from Index Team
Released by Index Team through its official repository on Hugging Face. Read the license.
Built From
- Derived from IndexTeam/Index-Echo-S2ST-9B
Questions About Index-Echo-S2ST-9B-FP8
Can I use Index-Echo-S2ST-9B-FP8 commercially?
Yes. Index-Echo-S2ST-9B-FP8 is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.
Similar Models
This is the model card of NLLB-200's distilled 600M variant. Here are the metrics for that particular checkpoint. - Information about training algorithms, parameters, fairness constraints or other applied approaches, and features. The exact training algorithm, data and the strategies to handle data imbalances for high and low resource languages that were used to train NLLB-200 is described in the paper. - Paper or other resource for more information NLLB Team et al, No Language Left Behind: Scaling Human-Centered Machine Translation, Arxiv, 2022 - Where to send questions or comments about the model: https://github.com/facebookresearch/fairseq/issues • Model performance measures: NLLB-200…
source languages: en; target languages: ru; OPUS readme: en-ru; dataset: opus; model: transformer-align; pre-processing: normalization + SentencePiece.
This model can be used for translation and text-to-text generation. CONTENT WARNING: Readers should be aware this section contains content that is disturbing, offensive, and can propagate historical and current stereotypes. Significant research has explored bias and fairness issues with language models (see, e.g., Sheng et al. (2021) and Bender et al. (2021)). Further details about the dataset for this model can be found in the OPUS readme: en-de
hfname: kor-eng - sourcelanguages: kor - targetlanguages: eng - opusreadmeurl: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/kor-eng/README.md - originalrepo: Tatoeba-Challenge - srcconstituents: {'korHani', 'korHang', 'korLatn', 'kor'} - tgtconstituents: {'eng'} - srcmultilingual: False - tgtmultilingual: False - urlmodel: https://object.pouta.csc.fi/Tatoeba-MT-models/kor-eng/opus-2020-06-17.zip - urltestset: https://object.pouta.csc.fi/Tatoeba-MT-models/kor-eng/opus-2020-06-17.test.txt - srcalpha3: kor - tgtalpha3: eng - shortpair: ko-en - chrF2score: 0.588 - brevitypenalty: 0.9590000000000001 - reflen: 17711.0 - srcname: Korean - tgtname: English - traindate…
source languages: de; target languages: en; OPUS readme: de-en; dataset: opus; model: transformer-align; pre-processing: normalization + SentencePiece.