{ "_id": "6a3d64cea319ef9414479b54", "Package": "chatterbox", "Title": "Text-to-Speech Using the 'Chatterbox' Engine", "Version": "0.2.1", "Date": "2026-06-21", "Authors@R": "c(person(\"Troy\", \"Hernandez\", role = c(\"aut\", \"cre\"),\nemail = \"troy@cornball.ai\",\ncomment = c(ORCID = \"0009-0005-4248-604X\")),\nperson(\"cornball.ai\", role = \"cph\"),\nperson(\"Resemble AI\", role = \"cph\",\ncomment = \"Chatterbox model architecture and weights (MIT license)\"))", "Description": "A native R 'torch' port of the 'Chatterbox' text-to-speech\nengine . Provides\nspeech synthesis with voice cloning; model weights are\ndownloaded from 'HuggingFace' via the\n'hfhub' package.", "License": "MIT + file LICENSE", "Encoding": "UTF-8", "URL": "https://github.com/cornball-ai/chatterbox", "BugReports": "https://github.com/cornball-ai/chatterbox/issues", "NeedsCompilation": "no", "Packaged": { "Date": "2026-06-25 17:21:18 UTC", "User": "root" }, "Author": "Troy Hernandez [aut, cre] (ORCID:\n), cornball.ai [cph],\nResemble AI [cph] (Chatterbox model architecture and weights\n(MIT license))", "Maintainer": "Troy Hernandez ", "Repository": "https://cran.r-universe.dev", "Date/Publication": "2026-06-25 11:50:18 UTC", "RemoteUrl": "https://github.com/cran/chatterbox", "RemoteRef": "HEAD", "RemoteSha": "e5342fd873e78dbf0d88548c3e50cee12e067bfc", "MD5sum": "a3ff3782e942fa58d76817bacd382e8e", "_user": "cran", "_type": "src", "_file": "chatterbox_0.2.1.tar.gz", "_fileid": "d23e361e2e8d1d08c6652c50cbd7cb5f0c18acdbb2dfc93af5f891c5e4971f8f", "_filesize": 1776901, "_sha256": "d23e361e2e8d1d08c6652c50cbd7cb5f0c18acdbb2dfc93af5f891c5e4971f8f", "_created": "2026-06-25T17:21:18.000Z", "_published": "2026-06-25T17:26:38.610Z", "_distro": "resolute", "_jobs": [ { "job": 83495291642, "time": 137, "config": "linux-devel-x86_64", "r": "4.7.0", "check": "OK", "artifact": "7885808034" }, { "job": 83495291638, "time": 128, "config": "linux-release-x86_64", "r": "4.6.0", "check": "OK", "artifact": "7885804260" }, { "job": 83494600298, "time": 213, "config": "source", "r": "4.6.0", "check": "OK", "artifact": "7885739714" }, { "job": 83495291584, "time": 147, "config": "wasm-release", "r": "4.6.0", "check": "OK", "artifact": "7885855251" } ], "_buildurl": "https://github.com/r-universe/cran/actions/runs/28187900601", "_status": "success", "_host": "GitHub-Actions", "_upstream": "https://github.com/cran/chatterbox", "_commit": { "id": "e5342fd873e78dbf0d88548c3e50cee12e067bfc", "author": "Troy Hernandez ", "committer": "cran-robot ", "message": "version 0.2.1\n", "time": 1782388218 }, "_maintainer": { "name": "Troy Hernandez", "email": "troy@cornball.ai", "login": "troyhernandez", "description": "", "uuid": 5670773 }, "_registered": true, "_dependencies": [ { "package": "R", "version": ">= 4.0.0", "role": "Depends" }, { "package": "torch", "version": ">= 0.17.0", "role": "Imports" }, { "package": "tuneR", "role": "Imports" }, { "package": "jsonlite", "role": "Imports" }, { "package": "hfhub", "role": "Imports" }, { "package": "safetensors", "version": ">= 0.1.1", "role": "Imports" }, { "package": "tinytest", "role": "Suggests" } ], "_owner": "cran", "_selfowned": false, "_usedby": 0, "_updates": [ { "week": "2026-26", "n": 1 } ], "_tags": [ { "name": "0.2.1", "date": "2026-06-25" } ], "_stars": 0, "_contributors": [ { "user": "troyhernandez", "count": 1, "uuid": 5670773 } ], "_userbio": { "uuid": 6899542, "type": "organization", "name": "cran", "followers": 611, "description": "Unofficial read-only mirror of all CRAN R packages" }, "_downloads": { "count": 0, "source": "https://cranlogs.r-pkg.org/downloads/total/last-month/chatterbox" }, "_devurl": "https://github.com/cornball-ai/chatterbox", "_searchresults": 10, "_rbuild": "4.6.0", "_assets": [ "extra/chatterbox.html", "extra/citation.cff", "extra/citation.html", "extra/citation.json", "extra/citation.txt", "extra/contents.json", "extra/NEWS.html", "extra/NEWS.txt", "LICENSE", "manual.pdf" ], "_cranurl": false, "_releases": [ { "version": "0.2.1", "date": "2026-06-25" } ], "_exports": [ "chatterbox", "chatterbox_gc_options", "create_voice_embedding", "download_chatterbox_models", "download_chatterbox_turbo_models", "generate", "generate_batch", "integrated_loudness", "load_chatterbox", "load_chatterbox_turbo", "load_voice_embedding", "models_available", "normalize_loudness", "normalize_tts_text", "quick_tts", "read_audio", "resample_audio", "s3_tokenizer", "save_voice_embedding", "serve", "tts_chunked", "tts_to_file", "turbo_models_available", "voice_convert", "write_audio" ], "_help": [ { "page": "apply_llama3_rope_scaling", "title": "Apply Llama3-style RoPE scaling", "topics": [ "apply_llama3_rope_scaling" ] }, { "page": "apply_rotary_emb_s3", "title": "Apply rotary position embeddings", "topics": [ "apply_rotary_emb_s3" ] }, { "page": "apply_rotary_pos_emb", "title": "Apply rotary position embeddings to Q and K", "topics": [ "apply_rotary_pos_emb" ] }, { "page": "attention_block", "title": "Attention block for perceiver", "topics": [ "attention_block" ] }, { "page": "basic_res_block", "title": "Basic residual block for FCM", "topics": [ "basic_res_block" ] }, { "page": "basic_transformer_block", "title": "Basic transformer block", "topics": [ "basic_transformer_block" ] }, { "page": "cam_dense_tdnn_block", "title": "CAM Dense TDNN Block (multiple layers with dense connections)", "topics": [ "cam_dense_tdnn_block" ] }, { "page": "cam_dense_tdnn_layer", "title": "CAM Dense TDNN Layer", "topics": [ "cam_dense_tdnn_layer" ] }, { "page": "cam_layer", "title": "CAM (Context-Aware Masking) Layer", "topics": [ "cam_layer" ] }, { "page": "campplus", "title": "CAMPPlus speaker encoder", "topics": [ "campplus" ] }, { "page": "causal_block1d", "title": "Causal Block 1D - CausalConv + LayerNorm + Mish", "topics": [ "causal_block1d" ] }, { "page": "causal_cfm", "title": "Causal Conditional Flow Matching", "topics": [ "causal_cfm" ] }, { "page": "causal_conv1d", "title": "Causal Conv1d - pads left only", "topics": [ "causal_conv1d" ] }, { "page": "causal_masked_diff_xvec", "title": "Causal Masked Diff with Xvector", "topics": [ "causal_masked_diff_xvec" ] }, { "page": "causal_resnet_block1d", "title": "Causal ResNet Block 1D", "topics": [ "causal_resnet_block1d" ] }, { "page": "cfm_attention", "title": "Self-attention for transformer block", "topics": [ "cfm_attention" ] }, { "page": "cfm_estimator", "title": "CFM Estimator (ConditionalDecoder)", "topics": [ "cfm_estimator" ] }, { "page": "chatterbox", "title": "Create (and load) a Chatterbox TTS model", "topics": [ "chatterbox" ] }, { "page": "chatterbox_gc_options", "title": "Recommended torch garbage-collection settings for chatterbox", "topics": [ "chatterbox_gc_options" ] }, { "page": "compute_rope_frequencies", "title": "Compute rotary position embeddings frequencies", "topics": [ "compute_rope_frequencies" ] }, { "page": "compute_ve_mel", "title": "Compute mel spectrogram for voice encoder", "topics": [ "compute_ve_mel" ] }, { "page": "conformer_encoder_layer", "title": "Conformer Encoder Layer", "topics": [ "conformer_encoder_layer" ] }, { "page": "conv_rnn_f0_predictor", "title": "Convolutional RNN F0 Predictor", "topics": [ "conv_rnn_f0_predictor" ] }, { "page": "create_kv_cache", "title": "Create pre-allocated KV cache", "topics": [ "create_kv_cache" ] }, { "page": "create_mel_filterbank", "title": "Create mel filterbank", "topics": [ "create_mel_filterbank" ] }, { "page": "create_voice_embedding", "title": "Create voice embedding from reference audio", "topics": [ "create_voice_embedding" ] }, { "page": "dense_layer", "title": "Dense layer for final embedding", "topics": [ "dense_layer" ] }, { "page": "download_chatterbox_models", "title": "Download Chatterbox Models from HuggingFace", "topics": [ "download_chatterbox_models" ] }, { "page": "download_chatterbox_turbo_models", "title": "Download Chatterbox Turbo Models from HuggingFace", "topics": [ "download_chatterbox_turbo_models" ] }, { "page": "drop_invalid_tokens", "title": "Drop invalid speech tokens", "topics": [ "drop_invalid_tokens" ] }, { "page": "espnet_rel_positional_encoding", "title": "Sinusoidal positional encoding (Espnet RelPositionalEncoding)", "topics": [ "espnet_rel_positional_encoding" ] }, { "page": "fcm_module", "title": "Factorized Convolutional Module (FCM)", "topics": [ "fcm_module" ] }, { "page": "feed_forward", "title": "Feed-forward network for transformer Matches diffusers FeedForward: net = [GELU(proj), Dropout, Linear]", "topics": [ "feed_forward" ] }, { "page": "fsmn_multi_head_attention", "title": "FSMN Multi-Head Attention", "topics": [ "fsmn_multi_head_attention" ] }, { "page": "fsq_codebook", "title": "FSQ Codebook module", "topics": [ "fsq_codebook" ] }, { "page": "fsq_vector_quantization", "title": "FSQ Vector Quantization wrapper", "topics": [ "fsq_vector_quantization" ] }, { "page": "gelu_with_proj", "title": "GELU activation with projection (matches diffusers GELU structure)", "topics": [ "gelu_with_proj" ] }, { "page": "generate", "title": "Generate speech from text", "topics": [ "generate" ] }, { "page": "generate_batch", "title": "Generate speech for several texts with one batched synthesis pass", "topics": [ "generate_batch" ] }, { "page": "get_conv_padding", "title": "Get padding for convolution", "topics": [ "get_conv_padding" ] }, { "page": "get_traced_layers", "title": "Get or create traced layers for cached inference", "topics": [ "get_traced_layers" ] }, { "page": "gpt2_attention", "title": "GPT-2 Attention (combined QKV projection)", "topics": [ "gpt2_attention" ] }, { "page": "gpt2_block", "title": "GPT-2 Transformer Block", "topics": [ "gpt2_block" ] }, { "page": "gpt2_config", "title": "GPT-2 Model Configuration", "topics": [ "gpt2_config" ] }, { "page": "gpt2_layer_norm", "title": "GPT-2 Layer Normalization", "topics": [ "gpt2_layer_norm" ] }, { "page": "gpt2_mlp", "title": "GPT-2 MLP (GELU activation)", "topics": [ "gpt2_mlp" ] }, { "page": "gpt2_model", "title": "GPT-2 Model (transformer backbone)", "topics": [ "gpt2_model" ] }, { "page": "hifigan_resblock", "title": "HiFiGAN Residual Block", "topics": [ "hifigan_resblock" ] }, { "page": "hift_generator", "title": "HiFTNet Generator", "topics": [ "hift_generator" ] }, { "page": "init_cache_from_first", "title": "Initialize cache with first token K/V values", "topics": [ "init_cache_from_first" ] }, { "page": "integrated_loudness", "title": "Integrated loudness (ITU-R BS.1770-4)", "topics": [ "integrated_loudness" ] }, { "page": "is_loaded", "title": "Check if model is loaded", "topics": [ "is_loaded" ] }, { "page": "learned_position_embeddings", "title": "Learned position embeddings module", "topics": [ "learned_position_embeddings" ] }, { "page": "linear_no_subsampling", "title": "Linear No Subsampling layer", "topics": [ "linear_no_subsampling" ] }, { "page": "llama_attention", "title": "Llama attention module", "topics": [ "llama_attention" ] }, { "page": "llama_config_520m", "title": "Create Llama 520M configuration", "topics": [ "llama_config_520m" ] }, { "page": "llama_decoder_layer", "title": "Llama decoder layer", "topics": [ "llama_decoder_layer" ] }, { "page": "llama_mlp", "title": "Llama MLP module", "topics": [ "llama_mlp" ] }, { "page": "llama_model", "title": "Llama model (decoder only)", "topics": [ "llama_model" ] }, { "page": "llama_rms_norm", "title": "RMS Normalization module", "topics": [ "llama_rms_norm" ] }, { "page": "load_chatterbox", "title": "Load Chatterbox model weights", "topics": [ "load_chatterbox" ] }, { "page": "load_chatterbox_turbo", "title": "Load Chatterbox Turbo model weights", "topics": [ "load_chatterbox_turbo" ] }, { "page": "load_conformer_encoder_weights", "title": "Load Conformer Encoder weights", "topics": [ "load_conformer_encoder_weights" ] }, { "page": "load_llama_weights", "title": "Load weights from safetensors into Llama model", "topics": [ "load_llama_weights" ] }, { "page": "load_t3_turbo_weights", "title": "Load T3 turbo weights from safetensors", "topics": [ "load_t3_turbo_weights" ] }, { "page": "load_t3_weights", "title": "Load T3 weights from safetensors", "topics": [ "load_t3_weights" ] }, { "page": "load_tokenizer", "title": "Load tokenizer from JSON file (internal)", "topics": [ "load_tokenizer" ] }, { "page": "load_voice_embedding", "title": "Load a voice embedding from disk", "topics": [ "load_voice_embedding" ] }, { "page": "load_voice_encoder_weights", "title": "Load voice encoder weights from safetensors", "topics": [ "load_voice_encoder_weights" ] }, { "page": "make_non_pad_mask_s3", "title": "Create non-padding mask", "topics": [ "make_non_pad_mask_s3" ] }, { "page": "make_pad_mask", "title": "Create padding mask", "topics": [ "make_pad_mask" ] }, { "page": "mask_to_bias", "title": "Convert mask to attention bias", "topics": [ "mask_to_bias" ] }, { "page": "mish_activation", "title": "Mish activation", "topics": [ "mish_activation" ] }, { "page": "models_available", "title": "Check if Models are Downloaded", "topics": [ "models_available" ] }, { "page": "normalize_loudness", "title": "Normalize audio to a target loudness", "topics": [ "normalize_loudness" ] }, { "page": "normalize_tts_text", "title": "Normalize text for TTS", "topics": [ "normalize_tts_text" ] }, { "page": "pad_audio_for_tokenizer", "title": "Pad audio to multiple of token rate", "topics": [ "pad_audio_for_tokenizer" ] }, { "page": "perceiver_resampler", "title": "Perceiver resampler for conditioning compression", "topics": [ "perceiver_resampler" ] }, { "page": "positionwise_feedforward", "title": "Positionwise Feed Forward", "topics": [ "positionwise_feedforward" ] }, { "page": "pre_lookahead_layer", "title": "Pre-Lookahead Layer", "topics": [ "pre_lookahead_layer" ] }, { "page": "precompute_freqs_cis", "title": "Precompute rotary position embedding frequencies", "topics": [ "precompute_freqs_cis" ] }, { "page": "print.chatterbox", "title": "Print method for chatterbox", "topics": [ "print.chatterbox" ] }, { "page": "print.chatterbox_gc_options", "title": "Print method for chatterbox_gc_options", "topics": [ "print.chatterbox_gc_options" ] }, { "page": "print.voice_embedding", "title": "Print method for voice_embedding", "topics": [ "print.voice_embedding" ] }, { "page": "punc_norm", "title": "Normalize punctuation for TTS", "topics": [ "punc_norm" ] }, { "page": "quick_tts", "title": "Quick TTS - one-line text-to-speech", "topics": [ "quick_tts" ] }, { "page": "read_audio", "title": "Read audio file", "topics": [ "read_audio" ] }, { "page": "reflection_pad1d", "title": "Reflection padding for 1D (nn_reflection_pad1d equivalent)", "topics": [ "reflection_pad1d" ] }, { "page": "rel_position_attention", "title": "Relative Position Multi-Headed Attention", "topics": [ "rel_position_attention" ] }, { "page": "resample_audio", "title": "Resample audio", "topics": [ "resample_audio" ] }, { "page": "rotate_half", "title": "Rotate half of the tensor for RoPE", "topics": [ "rotate_half" ] }, { "page": "s3_audio_encoder", "title": "S3 Audio Encoder V2", "topics": [ "s3_audio_encoder" ] }, { "page": "s3_log_mel_spectrogram", "title": "Compute log mel spectrogram for S3Tokenizer", "topics": [ "s3_log_mel_spectrogram" ] }, { "page": "s3_multi_head_attention", "title": "Multi-Head Attention base module", "topics": [ "s3_multi_head_attention" ] }, { "page": "s3_residual_attention_block", "title": "Residual attention block", "topics": [ "s3_residual_attention_block" ] }, { "page": "s3_tokenizer", "title": "S3Tokenizer V2 module", "topics": [ "s3_tokenizer" ] }, { "page": "s3_tokenizer_config", "title": "S3Tokenizer model configuration", "topics": [ "s3_tokenizer_config" ] }, { "page": "s3gen", "title": "S3Gen Token to Waveform", "topics": [ "s3gen" ] }, { "page": "save_voice_embedding", "title": "Save a voice embedding to disk", "topics": [ "save_voice_embedding" ] }, { "page": "serve", "title": "Serve chatterbox over HTTP", "topics": [ "serve" ] }, { "page": "sine_gen", "title": "Sine Generator", "topics": [ "sine_gen" ] }, { "page": "sinusoidal_pos_emb", "title": "Sinusoidal positional embedding for timesteps", "topics": [ "sinusoidal_pos_emb" ] }, { "page": "snake_activation", "title": "Snake activation function", "topics": [ "snake_activation" ] }, { "page": "source_module_hn_nsf", "title": "Source Module for Neural Source Filter", "topics": [ "source_module_hn_nsf" ] }, { "page": "statistics_pooling", "title": "Statistics pooling", "topics": [ "statistics_pooling" ] }, { "page": "t3_cond", "title": "Create T3 conditioning object", "topics": [ "t3_cond" ] }, { "page": "t3_cond_enc", "title": "T3 conditioning encoder", "topics": [ "t3_cond_enc" ] }, { "page": "t3_cond_to_device", "title": "Move T3 conditioning to device", "topics": [ "t3_cond_to_device" ] }, { "page": "t3_config_english", "title": "Create T3 configuration (English-only)", "topics": [ "t3_config_english" ] }, { "page": "t3_config_turbo", "title": "Create T3 turbo configuration (GPT-2 backbone)", "topics": [ "t3_config_turbo" ] }, { "page": "t3_inference_traced", "title": "T3 inference with JIT tracing (optimized)", "topics": [ "t3_inference_traced" ] }, { "page": "t3_model", "title": "T3 Token-to-Token TTS model", "topics": [ "t3_model" ] }, { "page": "t3_model_turbo", "title": "T3 Token-to-Token TTS model (Turbo variant with GPT-2 backbone)", "topics": [ "t3_model_turbo" ] }, { "page": "tdnn_layer", "title": "TDNN Layer", "topics": [ "tdnn_layer" ] }, { "page": "timestep_embedding", "title": "Timestep embedding MLP", "topics": [ "timestep_embedding" ] }, { "page": "tokenize_text", "title": "Encode text to token IDs using BPE", "topics": [ "tokenize_text" ] }, { "page": "traceable_attention", "title": "Traceable attention module with pre-allocated KV cache", "topics": [ "traceable_attention" ] }, { "page": "traceable_decoder_layer", "title": "Traceable decoder layer with pre-allocated KV cache", "topics": [ "traceable_decoder_layer" ] }, { "page": "traceable_kv_projector", "title": "Traceable K/V projection module", "topics": [ "traceable_kv_projector" ] }, { "page": "traceable_transformer_cached", "title": "Traceable transformer for cached inference", "topics": [ "traceable_transformer_cached" ] }, { "page": "traceable_transformer_first", "title": "Traceable transformer for first token (no cache)", "topics": [ "traceable_transformer_first" ] }, { "page": "transit_layer", "title": "Transit layer (channel reduction)", "topics": [ "transit_layer" ] }, { "page": "transpose_layer", "title": "Transpose layer for use in sequential", "topics": [ "transpose_layer" ] }, { "page": "tts_chunked", "title": "Generate speech for long text (the long-form policy layer)", "topics": [ "tts_chunked" ] }, { "page": "tts_to_file", "title": "Generate speech and save to file", "topics": [ "tts_to_file" ] }, { "page": "turbo_models_available", "title": "Check if Turbo Models are Downloaded", "topics": [ "turbo_models_available" ] }, { "page": "update_kv_cache", "title": "Update KV cache with new K/V values", "topics": [ "update_kv_cache" ] }, { "page": "update_valid_mask", "title": "Update valid mask to include new position", "topics": [ "update_valid_mask" ] }, { "page": "upsample_1d", "title": "Upsample 1D", "topics": [ "upsample_1d" ] }, { "page": "upsample_conformer_encoder", "title": "Upsample Conformer Encoder", "topics": [ "upsample_conformer_encoder" ] }, { "page": "upsample_conformer_encoder_full", "title": "Upsample Conformer Encoder", "topics": [ "upsample_conformer_encoder_full" ] }, { "page": "voice_convert", "title": "Convert speech to a target voice", "topics": [ "voice_convert" ] }, { "page": "voice_encoder", "title": "Voice encoder module", "topics": [ "voice_encoder" ] }, { "page": "voice_encoder_config", "title": "Voice encoder configuration", "topics": [ "voice_encoder_config" ] }, { "page": "write_audio", "title": "Write audio file", "topics": [ "write_audio" ] } ], "_rundeps": [ "askpass", "bit", "bit64", "callr", "cli", "coro", "curl", "desc", "farver", "filelock", "fs", "glue", "hfhub", "httr", "jsonlite", "labeling", "lifecycle", "magrittr", "MASS", "mime", "openssl", "otel", "processx", "ps", "R6", "RColorBrewer", "Rcpp", "rlang", "safetensors", "scales", "signal", "sys", "torch", "triebeard", "tuneR", "urltools", "viridisLite", "withr" ], "_score": 1, "_indexed": true, "_nocasepkg": "chatterbox", "_universes": [ "cran", "troyhernandez", "cornball-ai" ], "_binaries": [ { "r": "4.7.0", "os": "linux", "version": "0.2.1", "date": "2026-06-25T17:24:00.000Z", "distro": "resolute", "commit": "e5342fd873e78dbf0d88548c3e50cee12e067bfc", "fileid": "08aa08fdc85a7ab7a09632af740b549c3e7e0a12d0a6336a1964650c2389f18a", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/cran/actions/runs/28187900601" }, { "r": "4.6.0", "os": "linux", "version": "0.2.1", "date": "2026-06-25T17:23:50.000Z", "distro": "resolute", "commit": "e5342fd873e78dbf0d88548c3e50cee12e067bfc", "fileid": "f19d52168481911a4d3c5d98f89b219b77b0909da98e9f30ff764f9c25c9a64a", "status": "success", "check": "OK", "buildurl": "https://github.com/r-universe/cran/actions/runs/28187900601" }, { "r": "4.6.0", "os": "wasm", "version": "0.2.1", "date": "2026-06-25T17:26:18.000Z", "commit": "e5342fd873e78dbf0d88548c3e50cee12e067bfc", "fileid": "d319631335b5bedcd9fb7604575bd1bfae1bdddc2de31b373733e68a4a9c0297", "status": "success", "buildurl": "https://github.com/r-universe/cran/actions/runs/28187900601" } ] }