# Whisper

Whisper is a general-purpose speech recognition model with multilingual transcription, speech translation, and language identification capabilities.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## When Should a Voice Agent Speak? End-of-Turn Detection in Turkish

DevFeed: [When Should a Voice Agent Speak? End-of-Turn Detection in Turkish](<https://devfeed.tech/articles/when-should-a-voice-agent-speak-end-of-turn-detection-in-turkish-145522.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25489>)

Author: Ahmet Tu\\u{g}rul Bayrak, Fatma Nur Korkmaz, Mustafa Serta\\c{c} T\\"urkel, Nur Ya\\u{g}mur Tuncer

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Conversational AI](<https://devfeed.tech/topics/conversational-ai.md>), [speech-to-speech](<https://devfeed.tech/topics/speech-to-speech.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [detection](<https://devfeed.tech/tags/detection.md>), [end-of-turn-detection](<https://devfeed.tech/tags/end-of-turn-detection.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [features](<https://devfeed.tech/tags/features.md>), [fine](<https://devfeed.tech/tags/fine.md>), [language](<https://devfeed.tech/tags/language.md>), [language-model](<https://devfeed.tech/tags/language-model.md>), [latency](<https://devfeed.tech/tags/latency.md>), [logistic-regression](<https://devfeed.tech/tags/logistic-regression.md>), [model](<https://devfeed.tech/tags/model.md>), [recording](<https://devfeed.tech/tags/recording.md>), [recordings](<https://devfeed.tech/tags/recordings.md>), [response](<https://devfeed.tech/tags/response.md>), [simulation](<https://devfeed.tech/tags/simulation.md>), [speaker](<https://devfeed.tech/tags/speaker.md>), [study](<https://devfeed.tech/tags/study.md>), [tests](<https://devfeed.tech/tags/tests.md>), [text](<https://devfeed.tech/tags/text.md>), [voice](<https://devfeed.tech/tags/voice.md>)

### AI overview

In Turkish dyadic video calls, a fine-tuned Turkish TurnGPT model distinguished turn ends from mid-utterance pauses better than a zero-shot Turkish language model and a deployed multilingual detector, and was best calibrated among the three text systems. Current-speaker voice-activity timing was similarly informative; combining timing with the fine-tuned text score produced the strongest timing-based results, including against combinations using either text baseline with logistic regression. In a latency simulation, adding the text score to a silence timeout reduced mean response delay while keeping the interruption rate unchanged. The fine-tuned model also retained its advantage over the zero-shot model on automatic Whisper transcripts. Evaluation used recording-disjoint and speaker-disjoint splits.

### Source excerpt

arXiv:2611.25489v1 Announce Type: new Abstract: A voice agent that waits for a fixed silence before replying either interrupts users who pause mid-utterance or responds late to users who have finished. This study examines the decision at the onset of silence, using turn ends and turn-internal pauses extracted from the Real-TurnTurk corpus of Turkish dyadic video calls. A fine-tuned Turkish TurnGPT model, a zero-shot Turkish language model and a deployed multilingual detector are compared, together with voice-activity timing and facial features, under recording-disjoint and speaker-disjoint evaluation with paired cluster bootstrap tests. On held-out recordings, the fine-tuned model separates turn ends from pauses better than both text baselines and is the best calibrated of the three. Timing features of the current speaker are similarly informative, and their combination with the fine-tuned text score gives the best timing-based results; with logistic regression, this combination also outperforms the same combination with either text baseline. In a latency simulation, adding the text score to a silence timeout shortens the mean response delay at the same interruption rate. On automatic Whisper transcripts, the fine-tuned model retains its advantage over the zero-shot model.

## Run an LLM and Whisper on the Phone, With No Server Bill

DevFeed: [Run an LLM and Whisper on the Phone, With No Server Bill](<https://devfeed.tech/articles/run-an-llm-and-whisper-on-the-phone-with-no-server-bill-146235.md>)

Original publisher: [Read original article](<https://www.codenameone.com/blog/run-an-llm-on-the-phone/>)

Author: Shai Almog

Published: 2026-11-19T00:00:00Z

Content type: tutorial

Language: en

Sources: [CodeName One](<https://devfeed.tech/sources/codename-one.md>)

Topics: [Local AI](<https://devfeed.tech/topics/local-ai.md>), [ai-on-mobile](<https://devfeed.tech/topics/ai-on-mobile.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [llama.cpp](<https://devfeed.tech/topics/llama-cpp.md>)

Tags: [api](<https://devfeed.tech/tags/api.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [device](<https://devfeed.tech/tags/device.md>), [gguf](<https://devfeed.tech/tags/gguf.md>), [llm](<https://devfeed.tech/tags/llm.md>), [model](<https://devfeed.tech/tags/model.md>), [offline](<https://devfeed.tech/tags/offline.md>), [phone](<https://devfeed.tech/tags/phone.md>), [whisper](<https://devfeed.tech/tags/whisper.md>)

### AI overview

Codename One adds on-device LLM and speech transcription options that avoid per-request cloud charges and keep user text on the device. `LocalLlmClient` runs instruction-tuned GGUF models through the existing `LlmClient` API, supporting chat, streaming, embeddings, and grammar-enforced JSON; requests run sequentially on a background thread, with callbacks on the EDT. It works on Android and iOS, using Metal GPU layers on iOS, and can use Ollama in the simulator. The local client does not support tools or image inputs. The `cn1-ai-whisper` library runs whisper.cpp on Android, iOS, and the simulator; it accepts a ggml model and 16 kHz mono WAV, and can return timestamped transcript segments. Models require storage planning: the cited 0.5B, 4-bit model is about 400 MB and Whisper base about 150 MB, so the article recommends downloading them on first use. Native libraries are linked only when referenced.

### Source excerpt

Run a GGUF model on the phone with the same API as the cloud LLM clients, and transcribe speech offline with whisper. No tokens to pay for, and nothing leaves the device.

## Dictate into any app on Fedora with TuxTalk

DevFeed: [Dictate into any app on Fedora with TuxTalk](<https://devfeed.tech/articles/dictate-into-any-app-on-fedora-with-tuxtalk-135383.md>)

Original publisher: [Read original article](<https://fedoramagazine.org/dictate-into-any-app-on-fedora-with-tuxtalk/>)

Author: Ron Brand

Published: 2026-11-16T08:00:00Z

Content type: tutorial

Language: en

Sources: [Fedora Magazine](<https://devfeed.tech/sources/fedora-magazine.md>)

Topics: [Tool](<https://devfeed.tech/topics/tool.md>), [Fedora](<https://devfeed.tech/topics/fedora.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>), [keyboard](<https://devfeed.tech/topics/keyboard.md>)

Tags: [account](<https://devfeed.tech/tags/account.md>), [app](<https://devfeed.tech/tags/app.md>), [article](<https://devfeed.tech/tags/article.md>), [atomic](<https://devfeed.tech/tags/atomic.md>), [chat](<https://devfeed.tech/tags/chat.md>), [computer](<https://devfeed.tech/tags/computer.md>), [cursor](<https://devfeed.tech/tags/cursor.md>), [desktop](<https://devfeed.tech/tags/desktop.md>), [desktops](<https://devfeed.tech/tags/desktops.md>), [email](<https://devfeed.tech/tags/email.md>), [fedora](<https://devfeed.tech/tags/fedora.md>), [fedora-project-community](<https://devfeed.tech/tags/fedora-project-community.md>), [free](<https://devfeed.tech/tags/free.md>), [key](<https://devfeed.tech/tags/key.md>), [linux](<https://devfeed.tech/tags/linux.md>), [model](<https://devfeed.tech/tags/model.md>), [speech](<https://devfeed.tech/tags/speech.md>), [talk](<https://devfeed.tech/tags/talk.md>), [terminal](<https://devfeed.tech/tags/terminal.md>), [voice](<https://devfeed.tech/tags/voice.md>), [voice-dictation](<https://devfeed.tech/tags/voice-dictation.md>)

### AI overview

TuxTalk provides offline voice dictation on Fedora: hold F8 while speaking, or use F9 to toggle recording, and dictated text is entered at the cursor. Its first run downloads either the Parakeet model for English and 24 other European languages or Whisper for 99 languages. Installation uses a checksum-verified RPM on Fedora Workstation or KDE, and an AppImage on Atomic desktops; no GPU is required, though NVIDIA, AMD, and Intel graphics can use Vulkan. Setup adds a restricted systemd hotkey helper and a udev rule for virtual keyboard input. The excerpt also describes punctuation and email formatting, custom vocabulary, hands-free stopping, optional voice desktop controls, and uninstall options. Removing through TuxTalk's preferences removes its components and models; removing only the RPM leaves settings and downloaded models in the home folder.

### Source excerpt

TuxTalk is a free voice dictation tool for Linux. You hold a key, talk, and let go. What you say appears wherever your cursor is located: a document, an email, a chat box or a terminal. The speech model runs on your own computer, so your voice never leaves it and there is no account [...]

## F5-TTS voice cloning: how to clone your voice with AI on Runpod

DevFeed: [F5-TTS voice cloning: how to clone your voice with AI on Runpod](<https://devfeed.tech/articles/f5-tts-voice-cloning-how-to-clone-your-voice-with-ai-on-runpod-131264.md>)

Original publisher: [Read original article](<https://www.runpod.io/blog/clone-your-voice-f5-tts>)

Author: Daniel Monedero-Tortola

Published: 2026-11-12T23:13:09Z

Content type: tutorial

Language: en

Sources: [Runpod Blog.](<https://devfeed.tech/sources/runpod-blog.md>)

Topics: [Learning](<https://devfeed.tech/topics/learning.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [checkpoint](<https://devfeed.tech/tags/checkpoint.md>), [clone](<https://devfeed.tech/tags/clone.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [f5-tts](<https://devfeed.tech/tags/f5-tts.md>), [generate](<https://devfeed.tech/tags/generate.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [license](<https://devfeed.tech/tags/license.md>), [model](<https://devfeed.tech/tags/model.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [pod](<https://devfeed.tech/tags/pod.md>), [recording](<https://devfeed.tech/tags/recording.md>), [recordings](<https://devfeed.tech/tags/recordings.md>), [selection](<https://devfeed.tech/tags/selection.md>), [speech](<https://devfeed.tech/tags/speech.md>), [tts](<https://devfeed.tech/tags/tts.md>), [tts-voice-cloning](<https://devfeed.tech/tags/tts-voice-cloning.md>), [voice](<https://devfeed.tech/tags/voice.md>), [voice-cloning](<https://devfeed.tech/tags/voice-cloning.md>)

### AI overview

The tutorial explains how to fine-tune F5-TTS on at least 30 minutes of your own speech using a Runpod GPU pod, then generate speech from text in your voice. It covers recording matched audio and transcripts, preparing 24 kHz clips, training and comparing checkpoints, choosing reference audio, and downloading the model. It recommends a GPU with at least 24 GB VRAM and a persistent 50 GB network volume; training 30 minutes of audio for 200 epochs reportedly took about 75 minutes on an L4. Listen to samples and use an earlier checkpoint if later output becomes metallic from overtraining. The F5-TTS code is MIT-licensed, but its pretrained weights and fine-tuned derivatives are non-commercial; the tutorial suggests Chatterbox for commercial use. Keep the checkpoint private, since it can produce audio resembling your voice.

### Source excerpt

Fine-tune the open-source F5-TTS model on your own recordings and generate speech in your voice on a Runpod GPU pod. Covers recording, dataset prep, training, checkpoint selection, and the F5-TTS license.

## Removing Speech, Keeping Activities: Speech Removal for Privacy-Preserving Acoustic Sensing in Assisted Living

DevFeed: [Removing Speech, Keeping Activities: Speech Removal for Privacy-Preserving Acoustic Sensing in Assisted Living](<https://devfeed.tech/articles/removing-speech-keeping-activities-speech-removal-for-privacy-preserving-acoustic-sensing-in-assisted-living-114001.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2609.02376>)

Author: Pavlos Nicolaou, Christos Efstratiou

Published: 2026-11-03T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>)

Topics: [Speech Processing](<https://devfeed.tech/topics/speech-processing.md>), [fourier analysis](<https://devfeed.tech/topics/fourier-analysis.md>), [fourier transform](<https://devfeed.tech/topics/fourier-transform.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>), [asr](<https://devfeed.tech/topics/asr.md>), [Differential Privacy](<https://devfeed.tech/topics/differential-privacy.md>)

Tags: [acoustic-sensing](<https://devfeed.tech/tags/acoustic-sensing.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [audio](<https://devfeed.tech/tags/audio.md>), [cross-validation](<https://devfeed.tech/tags/cross-validation.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cs-hc](<https://devfeed.tech/tags/cs-hc.md>), [cs-sd](<https://devfeed.tech/tags/cs-sd.md>), [privacy](<https://devfeed.tech/tags/privacy.md>), [recognition](<https://devfeed.tech/tags/recognition.md>), [speech](<https://devfeed.tech/tags/speech.md>), [speech-recognition](<https://devfeed.tech/tags/speech-recognition.md>), [verification](<https://devfeed.tech/tags/verification.md>)

### AI overview

The paper presents a U-Net front end that masks the short-time Fourier transform to suppress speech while preserving sounds for activity recognition. It is trained on synthetic mixtures of public sounds and speech, avoiding the need to record speech in target homes. At a speech level 5 dB above background, activity-recognition macro F1 after removal is 0.574 and 0.661 on two datasets, versus 0.233 and 0.178 without processing; the method outperforms speech enhancement, speech separation, and detector-gated muting across tested speech levels. Whisper recovers 5.2% of spoken words after removal, versus 89.0% unprocessed, while voice-activity-detected speech on real home recordings drops from 8.5% to 2.7%. Speech and speaker identity are reduced, not eliminated.

### Source excerpt

arXiv:2609.02376v2 Announce Type: replace-cross Abstract: Acoustic sensing can monitor the daily activities of older adults without cameras or wearables, yet the possibility of capturing private conversations remains a barrier to its deployment. The engineering application addressed in this paper is privacy-preserving acoustic monitoring of daily activities in assisted living. Our contribution in artificial intelligence is a speech-removal front end: a U-Net that masks the short-time Fourier transform of ambient audio and returns a waveform in which speech is suppressed and the sounds needed for activity recognition are kept. The model is trained only on synthetic mixtures of public sounds and speech, so no speech has to be recorded in the target homes. We evaluate it with group-disjoint cross-validation on environmental and domestic sound datasets mixed with reverberant speech, on speech from unseen corpora and on real home recordings, and measure privacy with three voice activity detectors, Whisper speech recognition and a speaker-verification attack. With speech 5 dB above the background, the macro-averaged F1 score of activity recognition is 0.574 and 0.661 after removal on the two datasets, compared with 0.233 and 0.178 without processing, and higher than that of speech enhancement, speech separation and detector-gated muting at every speech level. Whisper recovers 5.2% of the spoken words, compared with 89.0% without processing and a chance floor of 2.2%. On the real recordings, the speech proportion estimated by a voice activity detector falls from 8.5% to 2.7%. The method reduces, but does not eliminate, spoken content and speaker identity.

## Custom Voices With Qwen3-TTS

DevFeed: [Custom Voices With Qwen3-TTS](<https://devfeed.tech/articles/custom-voices-with-qwen3-tts-107650.md>)

Original publisher: [Read original article](<https://lucumr.pocoo.org/2026/10/29/qwen3-custom-voices/>)

Author: Armin Ronacher

Published: 2026-10-29T00:00:00Z

Content type: article

Language: en

Sources: [Armin Ronacher](<https://devfeed.tech/sources/armin-ronacher.md>)

Topics: [Whisper](<https://devfeed.tech/topics/whisper.md>), [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [audio](<https://devfeed.tech/tags/audio.md>), [custom-voices](<https://devfeed.tech/tags/custom-voices.md>), [embeddings](<https://devfeed.tech/tags/embeddings.md>), [instructions](<https://devfeed.tech/tags/instructions.md>), [qwen3](<https://devfeed.tech/tags/qwen3.md>), [tts](<https://devfeed.tech/tags/tts.md>)

### AI overview

Qwen3-TTS voice embeddings can be blended and, with a workaround, paired with CustomVoice instructions to make distinctive narrated voices. The article contrasts Base, which clones from short reference audio but has no documented instruction control, with CustomVoice's nine speakers and instruction support and VoiceDesign's text-described voices. The author averages embeddings to blend voices and moves them along an axis derived from designed male and female voices; baking a blend into audio makes it easier to reuse, though it does not reproduce the blend exactly. Instructions worked on cloned voices less reliably than on built-in speakers. For speaker recognition, Qwen3 embeddings failed with a conference recording as reference but achieved 98% on podcast windows with a 20-second self-introduction; ECAPA identified the speaker in 97% of windows using the conference clip.

### Source excerpt

With my kids I'm currently building a little dungeron crawler and we're trying to learn a bit about game development along the way. However given that we live in the modern times of increasing capabilities we wanted to get a narrator and voice over. But we also wanted it to not sound like everyone's standard AI voices so I looked into what options there are on the table. After some investigation I narrowed down to using Qwen3-TTS which uses voice embeddings for custom voices. With a bit of creativity you can also do more with those custom voices than appears initially. In case this is of interest to you, this post just covers what we learned and with some examples. You can probably point your agent to it and it can do something with it. What Qwen Offers Qwen3 is not the only TTS model in town, and it wans't even my first choice. We initially used Kokoro. However Qwen3 supports voice cloning and we wanted to have it feel a bit more unique. In practice it's a bit tricker though. Qwen3 ships three kinds of models: 1.7B-Base (and 0.6B): clones a voice from a few seconds of reference audio 1.7B-CustomVoice: comes with nine built-in speakers and instructions 1.7B-VoiceDesign: invents a voice from a text description The base model clones either in context or from the speaker embedding alone but there is no documented way to clone a voice and direct it with instructions. That means unlike the custom voice and voice design models you cannot say "speak sad" or "whisper" with a cloned base model. At present the official options for a custom voice are fine-tuning Base on a single speaker or to make a clip with VoiceDesign and clone it with Base. That is not all that bad, but it means instructions are gone. Playing With It For this blog post I took 28 seconds of me from my talk at PyAI Conf, cut between two pauses so that it starts and ends on a full sentence, and transcribed it with Whisper. The transcript is really only needed for baking but to compare against an already exist

## Group Relative Policy Optimisation Improves Multilingual Speech Recognition in Low-Resource Languages

DevFeed: [Group Relative Policy Optimisation Improves Multilingual Speech Recognition in Low-Resource Languages](<https://devfeed.tech/articles/group-relative-policy-optimisation-improves-multilingual-speech-recognition-in-low-resource-languages-98094.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.30663>)

Author: Tim Wenjie Hu, Evelyn Nafula Ouma, Benjamin Akera, John A. Quinn

Published: 2026-10-23T04:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [asr](<https://devfeed.tech/topics/asr.md>), [grpo](<https://devfeed.tech/topics/grpo.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>), [Reinforcement learning](<https://devfeed.tech/topics/reinforcement-learning.md>), [Speech Processing](<https://devfeed.tech/topics/speech-processing.md>)

Tags: [asr](<https://devfeed.tech/tags/asr.md>), [audio](<https://devfeed.tech/tags/audio.md>), [autoregressive](<https://devfeed.tech/tags/autoregressive.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [decoder](<https://devfeed.tech/tags/decoder.md>), [errors](<https://devfeed.tech/tags/errors.md>), [extend](<https://devfeed.tech/tags/extend.md>), [grpo](<https://devfeed.tech/tags/grpo.md>), [inference](<https://devfeed.tech/tags/inference.md>), [languages](<https://devfeed.tech/tags/languages.md>), [llm](<https://devfeed.tech/tags/llm.md>), [optimisation](<https://devfeed.tech/tags/optimisation.md>), [policy](<https://devfeed.tech/tags/policy.md>), [recognition](<https://devfeed.tech/tags/recognition.md>), [reinforcement-learning](<https://devfeed.tech/tags/reinforcement-learning.md>), [speech](<https://devfeed.tech/tags/speech.md>), [speech-recognition](<https://devfeed.tech/tags/speech-recognition.md>), [teacher-forcing](<https://devfeed.tech/tags/teacher-forcing.md>), [training](<https://devfeed.tech/tags/training.md>), [training-data](<https://devfeed.tech/tags/training-data.md>)

### AI overview

The paper applies Group Relative Policy Optimisation (GRPO) to Whisper-large-v3, a non-LLM encoder-decoder speech recognition model, to address exposure bias from teacher-forced training. Across 51 African languages, the reported benefits are greatest when training data is scarce or audio is noisy or degraded. The resulting model, SunflowerASR, is reported to achieve the best published macro-averaged word error rates on AfriVox-v2 (0.227) and SimbaBench (0.372), and is publicly available.

### Source excerpt

arXiv:2610.30663v1 Announce Type: new Abstract: Autoregressive encoder-decoder speech recognition models trained with teacher forcing can suffer from exposure bias, limiting their ability to recover from their own prediction errors at inference time. Recent work has shown that reinforcement learning with Group Relative Policy Optimisation (GRPO) improves speech recognition in major languages with LLM-based models. We extend this line of work to smaller, non-LLM encoder-decoder ASR models, using Whisper-large-v3 as a case study across 51 African languages. The benefits of this approach are most pronounced when training data is scarce or when input audio is noisy or degraded. Our resulting model, SunflowerASR, achieves the best published macro-averaged WER on the AfriVox-v2 (0.227) and SimbaBench (0.372) benchmarks, and is publicly available.

## A Coding Guide to Cactus Whistle: Does a 16.9 MB Speech Model Hold Up Against Whisper Base? Word Error Rate, Speed, Streaming, Timestamps, Keywords, and Speech-to-Tool Calls on a CPU

DevFeed: [A Coding Guide to Cactus Whistle: Does a 16.9 MB Speech Model Hold Up Against Whisper Base? Word Error Rate, Speed, Streaming, Timestamps, Keywords, and Speech-to-Tool Calls on a CPU](<https://devfeed.tech/articles/a-coding-guide-to-cactus-whistle-does-a-16-9-mb-speech-model-hold-up-against-whisper-base-word-error-rate-speed-streaming-timestamps-keywords-and-speech-to-tool-calls-on-94653.md>)

Original publisher: [Read original article](<https://www.marktechpost.com/2026/10/21/cactus-whistle-vs-whisper-base-speech-to-text-benchmark-cpu-guide/>)

Author: Sana Hassan

Published: 2026-10-22T04:17:37Z

Content type: tutorial

Language: en

Sources: [MarkTechPost](<https://devfeed.tech/sources/marktechpost.md>)

Topics: [Whisper](<https://devfeed.tech/topics/whisper.md>), [Streaming](<https://devfeed.tech/topics/streaming.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [22](<https://devfeed.tech/tags/22.md>), [3](<https://devfeed.tech/tags/3.md>), [account](<https://devfeed.tech/tags/account.md>), [account-balance](<https://devfeed.tech/tags/account-balance.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [audio-language-model](<https://devfeed.tech/tags/audio-language-model.md>), [base](<https://devfeed.tech/tags/base.md>), [compute](<https://devfeed.tech/tags/compute.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [editors-pick](<https://devfeed.tech/tags/editors-pick.md>), [error](<https://devfeed.tech/tags/error.md>), [first](<https://devfeed.tech/tags/first.md>), [for-devs](<https://devfeed.tech/tags/for-devs.md>), [language-model](<https://devfeed.tech/tags/language-model.md>), [latency](<https://devfeed.tech/tags/latency.md>), [model](<https://devfeed.tech/tags/model.md>), [speech](<https://devfeed.tech/tags/speech.md>), [speed](<https://devfeed.tech/tags/speed.md>), [staff](<https://devfeed.tech/tags/staff.md>), [streaming](<https://devfeed.tech/tags/streaming.md>), [technology](<https://devfeed.tech/tags/technology.md>), [timestamps](<https://devfeed.tech/tags/timestamps.md>), [tool](<https://devfeed.tech/tags/tool.md>), [tutorials](<https://devfeed.tech/tags/tutorials.md>), [voice-ai](<https://devfeed.tech/tags/voice-ai.md>), [whisper](<https://devfeed.tech/tags/whisper.md>), [word](<https://devfeed.tech/tags/word.md>), [word-error-rate](<https://devfeed.tech/tags/word-error-rate.md>)

### AI overview

In the tested setup, Whistle is a 16.9 MB CPU speech model with a speed advantage over Whisper base, but its accuracy depends on the comparison: on a small paired LibriSpeech test-other sample it significantly beat greedy Whisper base, while the gap against five-beam search was not significant. The speed advantage narrowed from 7.7x on 5-second audio to 2.8x on 30-second audio; peak memory was 124 MB versus 823 MB for Whisper base. A pinned 3.3.0 engine silently dropped the middle of longer-than-30-second audio, while a 3.2.0 build refused it, making explicit chunking necessary in these tests. Streaming re-transcribes recent chunks and commits words after two passes agree; median commit delay rose with chunk size, with no significant WER difference from utterance-by-utterance transcription in the tested sessions. Word probabilities tracked errors well, and keyword gains required supplying correct words; distractor keywords worsened WER. Results are limited by small samples, few speakers, one-run Apple M4 measurements, and synthetic joined sessions; full-split figures were quoted rather than measured.

### Source excerpt

Explore a comprehensive coding guide benchmarking Cactus Compute's 16.9 MB speech model against Whisper base. Learn about word error rates, CPU speed, streaming latency, and speech-to-tool integration. The post A Coding Guide to Cactus Whistle: Does a 16.9 MB Speech Model Hold Up Against Whisper Base? Word Error Rate, Speed, Streaming, Timestamps, Keywords, and Speech-to-Tool Calls on a CPU appeared first on MarkTechPost.

## ASR Personalization in Pathological Voice: Accuracy, Efficiency, and Duration Effects

DevFeed: [ASR Personalization in Pathological Voice: Accuracy, Efficiency, and Duration Effects](<https://devfeed.tech/articles/asr-personalization-in-pathological-voice-accuracy-efficiency-and-duration-effects-93327.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.24572>)

Author: Giuseppe Giaimis, Sergei Katkov, Muhammad Azfar Yaqub, Antonio Liotta

Published: 2026-10-21T04:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [asr](<https://devfeed.tech/topics/asr.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>), [lora](<https://devfeed.tech/topics/lora.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>)

Tags: [adapter](<https://devfeed.tech/tags/adapter.md>), [asr](<https://devfeed.tech/tags/asr.md>), [attention](<https://devfeed.tech/tags/attention.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-sd](<https://devfeed.tech/tags/cs-sd.md>), [eess-as](<https://devfeed.tech/tags/eess-as.md>), [effects](<https://devfeed.tech/tags/effects.md>), [efficiency](<https://devfeed.tech/tags/efficiency.md>), [lora](<https://devfeed.tech/tags/lora.md>), [personalization](<https://devfeed.tech/tags/personalization.md>), [voice](<https://devfeed.tech/tags/voice.md>), [whisper](<https://devfeed.tech/tags/whisper.md>)

### AI overview

In single-speaker adaptation of Whisper for pathological speech, scaling the model reduces recognition errors but does not remove speaker-specific mismatch. The study compares LoRA placement, model size, training efficiency, and utterance duration in data-scarce settings. Full-attention LoRA is reported to provide the most reliable accuracy and best time-accuracy trade-off. Gains depend on recording length: medium-duration utterances benefit most, while short and long ones remain challenging. The supplied excerpt does not provide quantitative results or further methodological detail.

### Source excerpt

arXiv:2610.24572v1 Announce Type: cross Abstract: Automatic Speech recognition (ASR) struggles with Pathological speech due to persistent speaker-specific mismatch. In clinical and accessibility scenarios, only limited target-speaker data is typically available, making full model retraining impractical. We present a component-level analysis of LoRA adaptation in Whisper, investigating how adapter placement interacts with the model size, training efficiency and utterance-duration in single-speaker settings. Our experimental comparative study shows that while scaling reduces errors, it does not eliminate mismatch. Furthermore, the full attention LoRa provides the most reliable accuracy, and the best time--accuracy trade-off. Improvements vary across utterance length, with medium-length recordings benefiting the most, while short and long utterances remain challenging. These findings highlight the importance of adaptation placement and duration-dependent evaluation, under challenging, data-scarce personalization scenarios.

## Code-Switching Detection via Language-Selective Dimensions in Speech Foundation Models

DevFeed: [Code-Switching Detection via Language-Selective Dimensions in Speech Foundation Models](<https://devfeed.tech/articles/code-switching-detection-via-language-selective-dimensions-in-speech-foundation-models-87264.md>)

Original publisher: [Read original article](<https://machinelearning.apple.com/research/code-switching-detection>)

Published: 2026-10-14T00:00:00Z

Content type: article

Language: en

Sources: [Apple Machine Learning Research](<https://devfeed.tech/sources/apple-machine-learning-research.md>)

Topics: [asr](<https://devfeed.tech/topics/asr.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>), [Internationalization (i18n)](<https://devfeed.tech/topics/i18n.md>)

Tags: [asr](<https://devfeed.tech/tags/asr.md>), [detection](<https://devfeed.tech/tags/detection.md>), [foundation-models](<https://devfeed.tech/tags/foundation-models.md>), [generation](<https://devfeed.tech/tags/generation.md>), [speech](<https://devfeed.tech/tags/speech.md>)

### AI overview

The paper presents a lightweight method for detecting code-switching directly from speech. It aggregates frame-level activations from language-selective dimensions in pretrained speech encoders, and reports that the Whisper-small encoder achieves a mean F1 of 0.798 across 14 language pairs in CS-FLEURS. The findings suggest that pretrained speech encoders contain information useful for identifying code-switched utterances and building more representative multilingual corpora.

### Source excerpt

Code-switching (CS), the alternation between languages within an utterance, is common in multilingual speech but underrepresented in pretraining corpora. Increasing CS coverage requires finding relevant utterances in large speech collections so that they can be included in training data. One way to find such utterances is to transcribe the speech and detect language alternation in the text, but this depends on obtaining accurate multilingual transcriptions. We propose a lightweight method that detects CS directly from speech. It uses frame-level language information in pretrained speech...

## Top 12 AI Live Translation Tools in 2026: Latency, Languages, Pricing

DevFeed: [Top 12 AI Live Translation Tools in 2026: Latency, Languages, Pricing](<https://devfeed.tech/articles/top-12-ai-live-translation-tools-in-2026-latency-languages-pricing-85255.md>)

Original publisher: [Read original article](<https://www.marktechpost.com/2026/10/12/top-12-ai-live-translation-tools-2026/>)

Author: Michal Sutter

Published: 2026-10-12T08:20:49Z

Content type: comparison

Language: en

Sources: [MarkTechPost](<https://devfeed.tech/sources/marktechpost.md>)

Topics: [speech-to-speech](<https://devfeed.tech/topics/speech-to-speech.md>), [machine-translation](<https://devfeed.tech/topics/machine-translation.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>), [deepl](<https://devfeed.tech/topics/deepl.md>)

Tags: [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [ai](<https://devfeed.tech/tags/ai.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [audio-language-model](<https://devfeed.tech/tags/audio-language-model.md>), [editors-pick](<https://devfeed.tech/tags/editors-pick.md>), [for-devs](<https://devfeed.tech/tags/for-devs.md>), [latency](<https://devfeed.tech/tags/latency.md>), [pricing](<https://devfeed.tech/tags/pricing.md>), [staff](<https://devfeed.tech/tags/staff.md>), [tech-news](<https://devfeed.tech/tags/tech-news.md>), [technology](<https://devfeed.tech/tags/technology.md>), [top](<https://devfeed.tech/tags/top.md>), [translation](<https://devfeed.tech/tags/translation.md>), [voice-ai](<https://devfeed.tech/tags/voice-ai.md>)

### AI overview

This comparison reviews 12 AI live translation tools for conversations, meetings, events and webinars, covering language support, latency, platforms and pricing. It notes that language counts are not directly comparable and recommends piloting with your own language pairs.

### Source excerpt

We checked 12 AI live translation tools, including Palabra.ai, Google Live Translate, DeepL Voice, Teams Interpreter and Wordly, against official pages for languages, latency, platforms and pricing. The post Top 12 AI Live Translation Tools in 2026: Latency, Languages, Pricing appeared first on MarkTechPost.

## TII Releases Falcon-ASR: 1.6B Arabic Speech Model Hits 20.92% WER

DevFeed: [TII Releases Falcon-ASR: 1.6B Arabic Speech Model Hits 20.92% WER](<https://devfeed.tech/articles/tii-releases-falcon-asr-1-6b-arabic-speech-model-hits-20-92-wer-85059.md>)

Original publisher: [Read original article](<https://www.marktechpost.com/2026/10/11/tii-falcon-asr-arabic-speech-recognition/>)

Author: Asif Razzaq

Published: 2026-10-12T03:16:09Z

Content type: news

Language: en

Sources: [MarkTechPost](<https://devfeed.tech/sources/marktechpost.md>)

Topics: [asr](<https://devfeed.tech/topics/asr.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>)

Tags: [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-shorts](<https://devfeed.tech/tags/ai-shorts.md>), [applications](<https://devfeed.tech/tags/applications.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [asr](<https://devfeed.tech/tags/asr.md>), [audio-language-model](<https://devfeed.tech/tags/audio-language-model.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [editors-pick](<https://devfeed.tech/tags/editors-pick.md>), [for-devs](<https://devfeed.tech/tags/for-devs.md>), [language-model](<https://devfeed.tech/tags/language-model.md>), [large-language-model](<https://devfeed.tech/tags/large-language-model.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [new-releases](<https://devfeed.tech/tags/new-releases.md>), [releases](<https://devfeed.tech/tags/releases.md>), [small-language-model](<https://devfeed.tech/tags/small-language-model.md>), [software-engineering](<https://devfeed.tech/tags/software-engineering.md>), [speech](<https://devfeed.tech/tags/speech.md>), [staff](<https://devfeed.tech/tags/staff.md>), [tech-news](<https://devfeed.tech/tags/tech-news.md>), [technology](<https://devfeed.tech/tags/technology.md>), [voice-ai](<https://devfeed.tech/tags/voice-ai.md>)

### AI overview

TII released Falcon-ASR, a 1.6B-parameter speech recognition model focused on Emirati and other Arabic speech that also supports English, French, Spanish, and Portuguese. TII reports a 20.92% average word error rate across six Arabic test sets. The model is currently available as a hosted demo; its weights and API have not been published.

### Source excerpt

TII has released Falcon-ASR, a 1.6B speech recognition model for Arabic with an Emirati focus. It reports 20.92% average WER on 6 Arabic test sets, but weights and an API are not yet available. The post TII Releases Falcon-ASR: 1.6B Arabic Speech Model Hits 20.92% WER appeared first on MarkTechPost.

## Faster-Whisper Serverless Endpoint Offers 2-4x Faster Transcription

DevFeed: [Faster-Whisper Serverless Endpoint Offers 2-4x Faster Transcription](<https://devfeed.tech/articles/faster-whisper-3x-cheaper-4x-faster-transcription-84640.md>)

Original publisher: [Read original article](<https://www.runpod.io/blog/faster-whisper-serverless-endpoint>)

Author: Brandon Ikeler

Published: 2026-10-10T23:21:42Z

Content type: release

Language: en

Sources: [Runpod Blog.](<https://devfeed.tech/sources/runpod-blog.md>)

Topics: [Whisper](<https://devfeed.tech/topics/whisper.md>), [litellm](<https://devfeed.tech/topics/litellm.md>)

Tags: [api](<https://devfeed.tech/tags/api.md>), [cost](<https://devfeed.tech/tags/cost.md>), [serverless](<https://devfeed.tech/tags/serverless.md>), [transcription](<https://devfeed.tech/tags/transcription.md>), [whisper](<https://devfeed.tech/tags/whisper.md>)

### AI overview

Runpod says its Faster-Whisper serverless endpoint transcribes audio 2-4 times faster than its original Whisper endpoint and charges by execution time. The article says the original endpoint will be deprecated. It also notes that the feature described has since been discontinued and points readers to community-built pod templates.

### Source excerpt

Runpod's new Faster-Whisper endpoint delivers 2-4x faster transcription speeds than the original Whisper API, at a fraction of the cost.

## Introducing Falcon-ASR, an Arabic and multilingual speech recognition model

DevFeed: [Introducing Falcon-ASR, an Arabic and multilingual speech recognition model](<https://devfeed.tech/articles/introducing-falcon-asr-77329.md>)

Original publisher: [Read original article](<https://huggingface.co/blog/tiiuae/falcon-asr>)

Author: Abdul Muneer; Rishabh Saraf; Ramesh Gundluru; Shamsa Hamad; Lepauloux; Hakim Hacid

Published: 2026-10-07T13:21:03Z

Content type: release

Language: en

Sources: [Hugging Face](<https://devfeed.tech/sources/hugging-face-blog.md>)

Topics: [asr](<https://devfeed.tech/topics/asr.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>)

Tags: [asr](<https://devfeed.tech/tags/asr.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [hugging-face](<https://devfeed.tech/tags/hugging-face.md>), [open-asr-leaderboard](<https://devfeed.tech/tags/open-asr-leaderboard.md>)

### AI overview

The Technology Innovation Institute introduces Falcon-ASR, a 1.6-billion-parameter speech recognition model focused on Arabic, especially the Emirati dialect. It also supports English, French, Spanish, and Portuguese, with word-level timestamps. The post reports a 20.92% average word error rate across six Arabic test sets and 22.73% on an internal Emirati evaluation, and describes comparisons, training data, and evaluation conditions.

### Source excerpt

A Blog post by Technology Innovation Institute on Hugging Face

## Microsoft targets ultra-realistic voice agents with its first streaming transcription model

DevFeed: [Microsoft targets ultra-realistic voice agents with its first streaming transcription model](<https://devfeed.tech/articles/microsoft-targets-ultra-realistic-voice-agents-with-its-first-streaming-transcription-model-63645.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/10/01/microsoft-targets-ultra-realistic-voice-agents-with-its-first-streaming-transcription-model/>)

Author: Mike Wheatley

Published: 2026-10-02T01:31:31Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [Whisper](<https://devfeed.tech/topics/whisper.md>), [Conversational AI](<https://devfeed.tech/topics/conversational-ai.md>), [speech-to-speech](<https://devfeed.tech/topics/speech-to-speech.md>), [Streaming](<https://devfeed.tech/topics/streaming.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-transcription](<https://devfeed.tech/tags/ai-transcription.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [mai-models](<https://devfeed.tech/tags/mai-models.md>), [mai-thinking-1](<https://devfeed.tech/tags/mai-thinking-1.md>), [mai-transcribe-2-streaming](<https://devfeed.tech/tags/mai-transcribe-2-streaming.md>), [mai-voice-2-1](<https://devfeed.tech/tags/mai-voice-2-1.md>), [mai-voice-2-1-flash](<https://devfeed.tech/tags/mai-voice-2-1-flash.md>), [microsoft](<https://devfeed.tech/tags/microsoft.md>), [microsoft-ai](<https://devfeed.tech/tags/microsoft-ai.md>), [mustafa-suleyman](<https://devfeed.tech/tags/mustafa-suleyman.md>), [news](<https://devfeed.tech/tags/news.md>), [streaming-model](<https://devfeed.tech/tags/streaming-model.md>), [streaming-transcription](<https://devfeed.tech/tags/streaming-transcription.md>), [text-to-speech](<https://devfeed.tech/tags/text-to-speech.md>), [the-latest](<https://devfeed.tech/tags/the-latest.md>), [transcription](<https://devfeed.tech/tags/transcription.md>), [voice-agents](<https://devfeed.tech/tags/voice-agents.md>)

### AI overview

Microsoft released MAI-Transcribe-2-Streaming, its first streaming transcription model, alongside two text-to-speech models. The transcription model accepts speech over WebSocket, updates transcripts as a person speaks, supports more than 60 languages, and is listed at 54 cents per audio hour. Microsoft says its first transcript hypotheses arrive in 320 milliseconds on average, with performance depending on network and reply-generation conditions. The releases target developers building voice agents, and the article describes how combining transcription, reasoning, and speech generation gives developers control over quality, latency, and cost.

### Source excerpt

Microsoft Corp. today expanded its MAI artificial intelligence model family with its first streaming transcription model, debuting alongside two others focused on text-to-speech. They're designed for developers who want to build voice agents that can listen to people's voices and reply instantly, similar to how humans talk to one another. The most consequential of the [...] The post Microsoft targets ultra-realistic voice agents with its first streaming transcription model appeared first on SiliconANGLE.

## DIY AI voice assistant with ESP32 and animated OLED eyes

DevFeed: [DIY AI voice assistant with ESP32 and animated OLED eyes](<https://devfeed.tech/articles/diy-ai-voice-assistant-with-esp32-and-animated-oled-eyes-62642.md>)

Original publisher: [Read original article](<https://blog.adafruit.com/2026/09/30/diy-ai-voice-assistant-with-esp32-and-animated-oled-eyes/>)

Author: Anne Barela

Published: 2026-09-30T20:37:33Z

Content type: article

Language: en

Sources: [Adafruit Industries](<https://devfeed.tech/sources/adafruit-industries-makers-hackers-artists-designers-and-engineers.md>)

Topics: [ESP32](<https://devfeed.tech/topics/esp32.md>), [speech-to-speech](<https://devfeed.tech/topics/speech-to-speech.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>), [ChatGPT](<https://devfeed.tech/topics/chatgpt.md>)

Tags: [adafruit](<https://devfeed.tech/tags/adafruit.md>), [ai](<https://devfeed.tech/tags/ai.md>), [arduino](<https://devfeed.tech/tags/arduino.md>), [arduino-ide](<https://devfeed.tech/tags/arduino-ide.md>), [audio](<https://devfeed.tech/tags/audio.md>), [blocking](<https://devfeed.tech/tags/blocking.md>), [displays](<https://devfeed.tech/tags/displays.md>), [esp32](<https://devfeed.tech/tags/esp32.md>), [espressif](<https://devfeed.tech/tags/espressif.md>), [freertos](<https://devfeed.tech/tags/freertos.md>), [oled](<https://devfeed.tech/tags/oled.md>), [sound](<https://devfeed.tech/tags/sound.md>), [speech](<https://devfeed.tech/tags/speech.md>), [speech-to-text](<https://devfeed.tech/tags/speech-to-text.md>), [text-to-speech](<https://devfeed.tech/tags/text-to-speech.md>), [voice](<https://devfeed.tech/tags/voice.md>), [voice-assistant](<https://devfeed.tech/tags/voice-assistant.md>), [whisper](<https://devfeed.tech/tags/whisper.md>)

### AI overview

This DIY voice assistant runs on an ESP32: it sends recorded speech to the OpenAI Whisper API for transcription, uses GPT to generate a reply, and streams text-to-speech audio to a speaker. Animated OLED eyes run in a separate FreeRTOS task from the audio and network pipeline, so blocking HTTP calls do not stall the display.

### Source excerpt

oeadmin at Open Electronics covers a voice assistant project by jayesh_nawani that runs entirely on an ESP32. An I2S microphone picks up speech, the clip is packaged as a WAV file and sent to the OpenAI Whisper API for transcription. GPT writes a reply and the TTS audio streams back out through a MAX98357A amplifier [...]

## Transcribe and translate audio files with Faster Whisper

DevFeed: [Transcribe and translate audio files with Faster Whisper](<https://devfeed.tech/articles/transcribe-and-translate-audio-files-with-faster-whisper-78159.md>)

Original publisher: [Read original article](<https://www.runpod.io/blog/transcribe-and-translate-audio-files-with-faster-whisper>)

Author: Eliot Cowley

Published: 2026-09-26T14:29:16Z

Content type: tutorial

Language: en

Sources: [Runpod Blog.](<https://devfeed.tech/sources/runpod-blog.md>)

Topics: [Serverless](<https://devfeed.tech/topics/serverless.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>), [asr](<https://devfeed.tech/topics/asr.md>), [podcasting](<https://devfeed.tech/topics/podcasting.md>)

Tags: [asr](<https://devfeed.tech/tags/asr.md>), [deploy](<https://devfeed.tech/tags/deploy.md>), [efficiency](<https://devfeed.tech/tags/efficiency.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [openai](<https://devfeed.tech/tags/openai.md>), [serverless](<https://devfeed.tech/tags/serverless.md>), [whisper](<https://devfeed.tech/tags/whisper.md>)

### AI overview

This tutorial explains how to deploy Faster Whisper on Runpod Serverless to transcribe and translate audio files. It describes the model's speed, memory, and cost benefits and demonstrates calling a serverless endpoint with a Python request.

### Source excerpt

Learn how to deploy OpenAI's Faster Whisper on Runpod Serverless to transcribe and translate audio up to four times faster and at a fraction of the cost.

## Speaker-labeled transcription with WhisperX on SageMaker AI

DevFeed: [Speaker-labeled transcription with WhisperX on SageMaker AI](<https://devfeed.tech/articles/speaker-labeled-transcription-with-whisperx-on-sagemaker-ai-59574.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/machine-learning/speaker-labeled-transcription-with-whisperx-on-sagemaker-ai/>)

Author: Ayush Sharma

Published: 2026-09-24T16:20:12Z

Content type: tutorial

Language: en

Sources: [Artificial Intelligence](<https://devfeed.tech/sources/artificial-intelligence.md>)

Topics: [asr](<https://devfeed.tech/topics/asr.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>), [Amazon Web Services](<https://devfeed.tech/topics/aws.md>), [Amazon SageMaker](<https://devfeed.tech/topics/amazon-sagemaker.md>), [container](<https://devfeed.tech/topics/container.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [model-deployment](<https://devfeed.tech/topics/model-deployment.md>), [Inference](<https://devfeed.tech/topics/inference.md>), [Amazon S3](<https://devfeed.tech/topics/amazon-s3.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>)

Tags: [advanced-300](<https://devfeed.tech/tags/advanced-300.md>), [ai](<https://devfeed.tech/tags/ai.md>), [amazon-sagemaker](<https://devfeed.tech/tags/amazon-sagemaker.md>), [amazon-sagemaker-ai](<https://devfeed.tech/tags/amazon-sagemaker-ai.md>), [aws](<https://devfeed.tech/tags/aws.md>), [container](<https://devfeed.tech/tags/container.md>), [cost](<https://devfeed.tech/tags/cost.md>), [deploy](<https://devfeed.tech/tags/deploy.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [real-time](<https://devfeed.tech/tags/real-time.md>), [scaling](<https://devfeed.tech/tags/scaling.md>), [technical-how-to](<https://devfeed.tech/tags/technical-how-to.md>), [transcription](<https://devfeed.tech/tags/transcription.md>), [whisper](<https://devfeed.tech/tags/whisper.md>)

### AI overview

This tutorial explains how to deploy WhisperX on Amazon SageMaker AI for speaker-labeled transcription with precise word-level timestamps. It covers real-time and asynchronous endpoints, GPU deployment, scaling, Amazon S3 setup, and cost controls.

### Source excerpt

The AWS WhisperX Deep Learning Container packages Whisper, wav2vec2 forced alignment, and speaker diarization into a GPU-ready image. Learn how to deploy it to Amazon SageMaker AI real-time and asynchronous endpoints for word-level, speaker-labeled transcription, plus the production details that matter: the GPU AMI pin, scaling, and cost controls.

## PolyMO is a privacy-focused virtual pet built with an ESP32 and local AI

DevFeed: [PolyMO is a privacy-focused virtual pet built with an ESP32 and local AI](<https://devfeed.tech/articles/virtual-pet-that-won-t-let-you-waste-your-life-scrolling-34922.md>)

Original publisher: [Read original article](<https://www.yankodesign.com/2026/09/16/virtual-pet-that-wont-let-you-waste-your-life-scrolling/>)

Author: Ida Torres

Published: 2026-09-17T00:30:05Z

Content type: article

Language: en

Sources: [Yanko Design](<https://devfeed.tech/sources/yanko-design.md>)

Topics: [ESP32](<https://devfeed.tech/topics/esp32.md>), [Hardware](<https://devfeed.tech/topics/hardware.md>), [Language models](<https://devfeed.tech/topics/language-models.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>), [3D](<https://devfeed.tech/topics/3d.md>), [voice ai](<https://devfeed.tech/topics/voice-ai.md>), [Android](<https://devfeed.tech/topics/android.md>), [Bluetooth](<https://devfeed.tech/topics/bluetooth.md>)

Tags: [diy](<https://devfeed.tech/tags/diy.md>), [gadgets](<https://devfeed.tech/tags/gadgets.md>), [gadgets-technology-diy-virtual-virtual-pet](<https://devfeed.tech/tags/gadgets-technology-diy-virtual-virtual-pet.md>), [technology](<https://devfeed.tech/tags/technology.md>), [virtual](<https://devfeed.tech/tags/virtual.md>), [virtual-pet](<https://devfeed.tech/tags/virtual-pet.md>)

### AI overview

PolyMO is a virtual pet designed to discourage excessive scrolling by becoming visibly ill when ignored. It combines an ESP32 device with an Android phone running local language, speech-recognition, and text-to-speech models, keeping conversations and pet behavior on the owner's hardware.

### Source excerpt

Virtual Pet That Won't Let You Waste Your Life Scrolling Most of us know the feeling. You pick up your phone to check one notification, and forty-five minutes later you're three hundred posts deep into...

## Transcription on OpenRouter

DevFeed: [Transcription on OpenRouter](<https://devfeed.tech/articles/transcription-on-openrouter-78671.md>)

Original publisher: [Read original article](<https://openrouter.ai/blog/tutorials/transcription-on-openrouter/>)

Author: OpenRouter

Published: 2026-07-22T00:00:00Z

Content type: tutorial

Language: en

Sources: [OpenRouter](<https://devfeed.tech/sources/openrouter.md>)

Topics: [Chat completions endpoints](<https://devfeed.tech/topics/chat-completions-endpoints.md>), [API keys](<https://devfeed.tech/topics/api-keys.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>), [generative-ai-tools](<https://devfeed.tech/topics/generative-ai-tools.md>), [Handlebars](<https://devfeed.tech/topics/handlebars.md>), [Model Routing](<https://devfeed.tech/topics/model-routing.md>), [Back end](<https://devfeed.tech/topics/backend.md>), [Linear](<https://devfeed.tech/topics/linear.md>), [litellm](<https://devfeed.tech/topics/litellm.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-api](<https://devfeed.tech/tags/ai-api.md>), [api](<https://devfeed.tech/tags/api.md>), [audio](<https://devfeed.tech/tags/audio.md>), [azure](<https://devfeed.tech/tags/azure.md>), [browser](<https://devfeed.tech/tags/browser.md>), [chat](<https://devfeed.tech/tags/chat.md>), [cost](<https://devfeed.tech/tags/cost.md>), [curl](<https://devfeed.tech/tags/curl.md>), [json](<https://devfeed.tech/tags/json.md>), [model](<https://devfeed.tech/tags/model.md>), [openrouter](<https://devfeed.tech/tags/openrouter.md>), [text](<https://devfeed.tech/tags/text.md>), [transcription](<https://devfeed.tech/tags/transcription.md>)

### AI overview

This tutorial explains how to transcribe audio through OpenRouter's `POST /api/v1/audio/transcriptions` endpoint. It covers base64 and multipart requests, Whisper and newer speech-to-text models, response usage and cost fields, provider routing, timestamps, pricing, and operational limits such as the upstream timeout and lack of built-in SRT/VTT output.

### Source excerpt

Send base64 audio to POST /api/v1/audio/transcriptions and get JSON text plus a usage object back, with the same API key you already use for chat. Here's the full request contract, the model families, and the limits to design around.

## AI Gateway now supports streaming transcription

DevFeed: [AI Gateway now supports streaming transcription](<https://devfeed.tech/articles/ai-gateway-now-supports-streaming-transcription-802.md>)

Original publisher: [Read original article](<https://vercel.com/changelog/ai-gateway-now-supports-streaming-transcription>)

Author: Jerilyn Zheng

Published: 2026-07-22T00:00:00Z

Content type: release

Language: en

Sources: [Vercel News](<https://devfeed.tech/sources/vercel-news.md>)

Topics: [Streaming](<https://devfeed.tech/topics/streaming.md>), [asr](<https://devfeed.tech/topics/asr.md>), [SDK](<https://devfeed.tech/topics/sdk.md>), [vercel ai sdk](<https://devfeed.tech/topics/vercel-ai-sdk.md>), [Latency](<https://devfeed.tech/topics/latency.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [gpt](<https://devfeed.tech/tags/gpt.md>), [latency](<https://devfeed.tech/tags/latency.md>), [openai](<https://devfeed.tech/tags/openai.md>), [sdk](<https://devfeed.tech/tags/sdk.md>), [speech](<https://devfeed.tech/tags/speech.md>), [streaming](<https://devfeed.tech/tags/streaming.md>), [streams](<https://devfeed.tech/tags/streams.md>), [transcription](<https://devfeed.tech/tags/transcription.md>), [transcripts](<https://devfeed.tech/tags/transcripts.md>), [voice](<https://devfeed.tech/tags/voice.md>), [whisper](<https://devfeed.tech/tags/whisper.md>)

### AI overview

AI Gateway now supports beta streaming transcription through the AI SDK's streamTranscribe function. Applications can stream audio as it is captured and receive partial and final transcript updates with low latency, enabling use cases such as live captioning, voice input, and voice-enabled agents.

### Source excerpt

AI Gateway now supports streaming transcription. Previously, transcription required a complete audio file and returned the full transcript in a single response. Now you can stream audio in as it's captured and get transcript updates back as the model produces them, keeping latency low for uses like live captioning and voice input. Streaming transcription is in beta and available through the AI SDK's streamTranscribe function with any streaming-capable transcription model. The example below streams raw PCM audio to openai/gpt-realtime-whisper and prints each transcript delta as it arrives. The result stream also carries partial and final transcripts: The same code works across providers: swap the model string to use xai/grok-stt or any other streaming-capable transcription model. Streaming transcription also makes it easy to add a voice mode to an agent. Stream the user's speech to a transcription model and pass the live text to your agent. The agent itself does not change: it still receives text, so this works with any text-based agent. For agents that speak back, pair it with speech generation, or use realtime voice for full two-way conversation. For more information on streaming transcription on AI Gateway, see the documentation. Read more

## VideoIO, PCM Mixing And Timed Whisper Captions

DevFeed: [VideoIO, PCM Mixing And Timed Whisper Captions](<https://devfeed.tech/articles/videoio-pcm-mixing-and-timed-whisper-captions-19662.md>)

Original publisher: [Read original article](<https://www.codenameone.com/blog/videoio-audio-mixer-whisper/>)

Author: Shai Almog

Published: 2026-07-08T00:00:00Z

Content type: release

Language: en

Sources: [CodeName One](<https://devfeed.tech/sources/codename-one.md>)

Topics: [API](<https://devfeed.tech/topics/api.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>), [cross-platform](<https://devfeed.tech/topics/cross-platform.md>)

Tags: [api](<https://devfeed.tech/tags/api.md>), [audio](<https://devfeed.tech/tags/audio.md>), [cross-platform](<https://devfeed.tech/tags/cross-platform.md>), [encoding](<https://devfeed.tech/tags/encoding.md>), [playback](<https://devfeed.tech/tags/playback.md>), [release](<https://devfeed.tech/tags/release.md>), [transcription](<https://devfeed.tech/tags/transcription.md>), [video](<https://devfeed.tech/tags/video.md>), [whisper](<https://devfeed.tech/tags/whisper.md>)

### AI overview

Codename One's release adds cross-platform video encoding and decoding, sample-accurate PCM mixing, and timed Whisper transcription segments with SRT and VTT subtitle output. It can encode app-rendered frames, decode exact frames and audio where supported, and package native integrations across several platforms.

### Source excerpt

Codename One can now generate and inspect real video: encode app-rendered frames, decode exact frames, mix PCM, and turn Whisper timestamps into subtitles.

## Meeting Notes - A Free Desktop App for Tracking 1:1s

DevFeed: [Meeting Notes - A Free Desktop App for Tracking 1:1s](<https://devfeed.tech/articles/meeting-notes-a-free-desktop-app-for-tracking-1-1s-37543.md>)

Original publisher: [Read original article](<https://deanhume.com/meeting-notes-desktop-app-windows/>)

Author: Dean Hume

Published: 2026-06-29T10:16:48Z

Content type: article

Language: en

Sources: [Dean Hume](<https://devfeed.tech/sources/dean-hume.md>)

Topics: [App](<https://devfeed.tech/topics/app.md>), [meetings](<https://devfeed.tech/topics/meetings.md>), [Local-First](<https://devfeed.tech/topics/local-first.md>), [asr](<https://devfeed.tech/topics/asr.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>), [Markdown](<https://devfeed.tech/topics/markdown.md>)

Tags: [app](<https://devfeed.tech/tags/app.md>), [autosave](<https://devfeed.tech/tags/autosave.md>), [desktop](<https://devfeed.tech/tags/desktop.md>), [free](<https://devfeed.tech/tags/free.md>), [meetings](<https://devfeed.tech/tags/meetings.md>), [notes](<https://devfeed.tech/tags/notes.md>), [offline](<https://devfeed.tech/tags/offline.md>), [on-device](<https://devfeed.tech/tags/on-device.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [recording](<https://devfeed.tech/tags/recording.md>), [speech-to-text](<https://devfeed.tech/tags/speech-to-text.md>), [technical-leadership](<https://devfeed.tech/tags/technical-leadership.md>), [technical-program-manager](<https://devfeed.tech/tags/technical-program-manager.md>), [whisper](<https://devfeed.tech/tags/whisper.md>), [writing](<https://devfeed.tech/tags/writing.md>)

### AI overview

The article introduces Meeting Notes, a free, open-source Windows desktop app for organizing 1:1 notes by person. It supports topic tags, Markdown, autosave, voice recording, and offline speech-to-text using Whisper, with data and audio kept on the device.

### Source excerpt

Meeting Notes is a free offline desktop app for 1:1s. Organise notes by person, tag topics, and transcribe meetings on-device with Whisper.

## Violin: An open-source video translation skill that breaks language barriers

DevFeed: [Violin: An open-source video translation skill that breaks language barriers](<https://devfeed.tech/articles/violin-an-open-source-video-translation-skill-that-breaks-language-barriers-80231.md>)

Original publisher: [Read original article](<https://www.together.ai/blog/violin-open-source-translation-skill>)

Author: Shang Zhu, Kevin Qinghong Lin (Oxford), James Zou

Published: 2026-05-14T00:00:00Z

Content type: release

Language: en

Sources: [Together.ai](<https://devfeed.tech/sources/together-ai.md>)

Topics: [Generative AI](<https://devfeed.tech/topics/generative-ai.md>), [speech-to-speech](<https://devfeed.tech/topics/speech-to-speech.md>), [Whisper](<https://devfeed.tech/topics/whisper.md>), [gui-agents](<https://devfeed.tech/topics/gui-agents.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [audio](<https://devfeed.tech/tags/audio.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [speech-recognition](<https://devfeed.tech/tags/speech-recognition.md>), [text-to-speech](<https://devfeed.tech/tags/text-to-speech.md>)

### AI overview

Violin is an open-source tool for translating videos through speech recognition, LLM translation and text-to-speech. It also offers video-aware chat, voice selection, a web app, a CLI for scripting and batch processing, and agent skills. The article describes the tool's pipeline and says its code is released under the MIT license.

### Source excerpt

Violin is an open-source AI video translation tool that combines speech recognition, LLM translation, and text-to-speech to make video content accessible across languages.

[Next page](<https://devfeed.tech/topics/whisper.md?cursor=WyIyMDI2LTA1LTE0VDAwOjAwOjAwKzAwOjAwIiwgIjllOWI5YWM0LWI2MDYtNGY0My05NDg4LTQzYmFhMDg2MTQ0MiJd>)