--- license: apache-2.0 language: - en library_name: transformers pipeline_tag: text-to-speech base_model: OpenMOSS-Team/MOSS-TTS-Nano tags: - text-to-speech - tts - moss-tts-nano - indian-english - lora - voice-agent - on-device model-index: - name: roxi-tts-v3.1 results: - task: type: text-to-speech name: Text-to-Speech metrics: - type: speaker-similarity name: Speaker similarity (WavLM-SV, vs target) value: 0.96 - type: wer name: Intelligibility WER (Whisper-base.en) value: 0.33 --- # šŸŽ™ļø Roxi-TTS v3.1 ā€” Indian-English voice (MOSS-TTS-Nano LoRA)

base license lang sr method

A compact **Indian-English** text-to-speech voice ā€” a LoRA fine-tune of the 0.1 B [**MOSS-TTS-Nano**](https://huggingface.co/OpenMOSS-Team/MOSS-TTS-Nano) on **~4 hours** of a single studio speaker. Built for the **VozVox** voice-agent platform (customer-support / website assistants). Tiny, fast, 48 kHz, and commercially permissive end-to-end. > **This is the current best of the Roxi line** (preferred by ear over v2 and v3). It's an honest > 0.1 B proof-of-concept: natural and clearly Indian, but read-speech in style ā€” not yet fully > conversational. See [Limitations](#-limitations). ## šŸ“‹ Model at a glance | | | |---|---| | **Base model** | `OpenMOSS-Team/MOSS-TTS-Nano` (0.1 B, autoregressive audio-token + LLM) ā€” Apache-2.0 | | **Audio tokenizer** | `OpenMOSS-Team/MOSS-Audio-Tokenizer-Nano` (Apache-2.0) | | **Method** | LoRA (PEFT) ā€” r=32, Ī±=64, targets `c_attn, c_proj, fc_in, fc_out` (~4.2% params), BF16, merged | | **Training data** | ~4 h, single IndicTTS-English speaker (studio, 48 kHz), 2,634 clips | | **Output** | 48 kHz mono WAV | | **Speaker similarity** | **0.96** (WavLM-SV cosine to held-out target) | | **Intelligibility (WER)** | **0.33** (Whisper-base.en on generated audio) | ## šŸ§¬ The Roxi line | Model | Speaker / data | Speaker-sim | WER | Notes | |---|---|---|---|---| | [`roxi-tts-v2`](https://huggingface.co/IOTEverythin/roxi-tts-v2) | speaker A, ~50 min, r16 | 0.96 | 0.26 | milder voice | | [`roxi-tts-v3`](https://huggingface.co/IOTEverythin/roxi-tts-v3) | speaker B, ~70 min, r16 | 0.96 | 0.29 | different voice, fewer cut-offs | | **`roxi-tts-v3.1`** (this) | speaker B, **~4 h**, r32 | 0.96 | 0.33 | **preferred by ear (smoothest)** | | [`roxi-tts-v2-onnx`](https://huggingface.co/IOTEverythin/roxi-tts-v2-onnx) | ONNX/CPU build of v2 | 0.73 | 0.25 | **no transformers dependency** | ## āš ļø Requirements ā€” please read This model uses MOSS-TTS-Nano's **custom code (`trust_remote_code`)**, which is built for **`transformers==4.57.1`**. On transformers 5.x it produces NaN/noise. Pin it, and **restart your runtime** after installing (Colab preloads 5.x): ```bash pip install "transformers==4.57.1" torch torchaudio soundfile librosa sentencepiece ``` ```python import transformers; assert transformers.__version__ == "4.57.1" # verify before generating ``` Also: **load one `trust_remote_code` model per kernel** (loading two corrupts the module cache). ## šŸš€ Usage ```python import torch from transformers import AutoModelForCausalLM device = "cuda" if torch.cuda.is_available() else "cpu" model = AutoModelForCausalLM.from_pretrained( "IOTEverythin/roxi-tts-v3.1", trust_remote_code=True, dtype=torch.float32 ).to(device).eval() res = model.inference( text="Welcome. Your appointment is confirmed for Monday at ten thirty in the morning.", output_audio_path="out.wav", mode="continuation", audio_tokenizer_type="moss-audio-tokenizer-nano", audio_tokenizer_pretrained_name_or_path="OpenMOSS-Team/MOSS-Audio-Tokenizer-Nano", device=device, audio_repetition_penalty=1.1, use_kv_cache=True, ) from IPython.display import Audio; Audio("out.wav") ``` ### Recommended helper (retry + trim) Generation is autoregressive and occasionally under-generates (cuts off) ā€” retry and trim silence: ```python import numpy as np, soundfile as sf, librosa from IPython.display import Audio, display def say(text, tries=6): target = len(text.split())/3.0; best=(0,None,24000) for _ in range(tries): model.inference(text=text, output_audio_path="out.wav", mode="continuation", audio_tokenizer_type="moss-audio-tokenizer-nano", audio_tokenizer_pretrained_name_or_path="OpenMOSS-Team/MOSS-Audio-Tokenizer-Nano", device=device, audio_repetition_penalty=1.1, use_kv_cache=True) y,sr = sf.read("out.wav"); y = y.mean(1) if y.ndim>1 else y yt,_ = librosa.effects.trim(y.astype(np.float32), top_db=35) if len(yt)/sr > best[0]: best=(len(yt)/sr, yt, sr) if len(yt)/sr >= target*0.9: break display(Audio(best[1], rate=best[2])) say("Our Bengaluru office is open until six thirty this evening.") ``` **Tips:** spell brands phonetically ("Voz Vox"), avoid raw abbreviations ("in the morning", not "A M"), write numbers as words, and keep sentences ā‰¤ ~12 words for reliability. Do **not** raise `max_new_frames` (the codec decode is O(nĀ²) memory and can OOM). ## šŸŽÆ Intended use Indian-English TTS for **customer-support calls** and **website voice assistants** ā€” natural, conversational, warm/professional, telephony-aware. Single-speaker branded voice. ## šŸ—ļø Training - Data: a single expressive (storytelling) speaker isolated from `SPRINGLab/IndicTTS-English` via WavLM-SV speaker clustering across dataset shards (~2,634 clips, ~4 h, studio 48 kHz, proper-case transcripts). - LoRA r=32/Ī±=64 on `c_attn, c_proj, fc_in, fc_out`, BF16, lr 1e-4 cosine, grad-accum 8; **epoch-2** checkpoint selected (later epochs overfit the reading style and raised WER). - Evaluation on held-out clips: speaker similarity via `microsoft/wavlm-base-plus-sv`; intelligibility via `openai/whisper-base.en` on the actual generated audio. ## šŸ§± Limitations - **0.1 B model** ā€” sounds synthetic vs larger TTS; naturalness is capped by size. - **Read-speech data** ā€” delivery is somewhat formal, not fully conversational; accent is the training speaker's (clear, mildly Indian), not strongly stereotypical. - **Stochastic cut-offs** ā€” use the retry helper; keep sentences short. - **Telephony (8 kHz)** not separately tuned. **Style/emotion control** is not reliable (neutral only). - Requires `transformers==4.57.1` (see Requirements). ## šŸ™ Attribution & license Released under **Apache-2.0**. Built on **MOSS-TTS-Nano** (Apache-2.0) and its audio tokenizer (Apache-2.0). Training data: IIT-Madras **Indic TTS** (English) via `SPRINGLab/IndicTTS-English`. Required notice: > COPYRIGHT 2016 TTS Consortium, TDIL, Meity ā€” represented by Hema A. Murthy & S. Umesh, > Department of Computer Science and Engineering and Electrical Engineering, IIT Madras. ALL RIGHTS RESERVED. ## šŸ›”ļø Responsible use This voice is derived from a real dataset speaker. **Do not** use it to impersonate real people, for fraud, social engineering, or deception. Disclose AI-generated audio where required by law/policy. Provided "as is", without warranty.