Los mejores modelos de voz a texto y transcripción
Clasificaciones de modelos actualizadas según datos de uso reales.
Los modelos se clasifican por el total de caracteres transcritos procesados a través de la API de OpenRouter durante los últimos 7 días. Se muestran los 25 modelos mejor clasificados. Las clasificaciones miden el uso en OpenRouter y reflejan la adopción, no la calidad del modelo ni su rendimiento en pruebas de evaluación.
Los modelos de voz a texto convierten el audio hablado en texto para transcripciones, subtítulos, resúmenes de reuniones, análisis de llamadas y aplicaciones controladas por voz. Esta colección clasifica los modelos de transcripción según su uso en OpenRouter durante la última semana. Los mejores modelos actuales son Whisper Large V3 Turbo, MAI-Transcribe 2 y Whisper Large V3. Compare la precisión, la velocidad, los idiomas admitidos y el costo para encontrar el modelo adecuado para su flujo de trabajo de audio.
Mejores modelos de voz a texto en OpenRouter
OpenAI: Whisper Large V3 Turbo
Whisper Large V3 Turbo is an optimized version of OpenAI's Whisper Large V3 speech recognition model, designed for speed and cost efficiency. It supports transcription across 99+ languages with a 12% word error rate, and accepts common audio formats including mp3, mp4, wav, webm, flac, and ogg. Achieves real-time speed factors up to 216x, making it well-suited for latency-sensitive and high-throughput transcription workloads.

Microsoft AI: MAI-Transcribe 2
MAI-Transcribe 2 is a multilingual speech-to-text model from Microsoft AI, ranked #1 on the FLEURS multilingual benchmark. It supports 60 languages with automatic language identification, code switching for mixed-language speech, speaker diarization, word-level timestamps, keyword biasing for domain-specific terminology, and configurable verbatim or clean transcription styles. It is suited for captions, call transcription, subtitling, accessibility, and other voice-enabled applications, and is faster than MAI-Transcribe-1.5 on long-form audio.
On OpenRouter, set response_format to "verbose_json" for segment timestamps, and add timestamp_granularities: ["word"] for word-level timestamps. Set provider.options.azure.diarization.enabled to true for speaker labels, provide keywords through provider.options.azure.phraseList.phrases, and select a transcription style through provider.options.azure.enhancedMode.modelOptions.transcribeStyle. See the speech-to-text guide.
OpenAI: Whisper Large V3
Whisper Large V3 is OpenAI's open-source automatic speech recognition model offering both audio transcription and translation. It supports 99+ languages and accepts common audio formats including mp3, mp4, wav, webm, flac, and ogg. With 1,550M parameters, it achieves a 10.3% word error rate and is well-suited for noise-robust, multilingual transcription in demanding conditions. Supports timestamp granularities at word and segment levels.
ElevenLabs: Scribe v2
ElevenLabs Scribe v2 is a speech-to-text model that transcribes audio in 90+ languages with word-level timestamps, optional speaker diarization, and audio-event tagging.
OpenAI: GPT-4o Mini Transcribe
GPT-4o Mini Transcribe is OpenAI's smaller, cost-efficient speech-to-text model built on GPT-4o Mini audio capabilities. It's priced per token (input and output), making it suitable for high-volume transcription workflows that benefit from token-level billing transparency at a lower cost point.

Qwen: Qwen3 ASR 1.7B
Qwen3 ASR 1.7B is an automatic speech recognition model from Qwen. It supports multilingual language identification and transcription across 30 languages and 22 Chinese dialects, with streaming and offline inference plus segment-level and word-level timestamps.
OpenAI: Whisper 1
Whisper is OpenAI's open-source automatic speech recognition model, available via API as whisper-1. It supports transcription and translation across 50+ languages from audio files up to 25 MB. Accepts formats including mp3, mp4, wav, and webm. Priced per minute of audio duration, billed to the nearest second.
AssemblyAI: Universal-3.6 Pro
Universal-3.6 Pro is AssemblyAI's speech-to-text model and the successor to Universal-3.5 Pro, served here through AssemblyAI's Sync API. Compared with Universal-3.5 Pro, AssemblyAI reports 30% fewer errors on short conversational utterances, 12% fewer errors on average across supported languages, 14 newly supported languages, and better recognition and formatting of entities such as phone numbers, addresses, emails, and codes. It returns a complete transcript with word-level timestamps in a single synchronous response for audio up to 20 minutes, accepts WAV and compressed input (MP3, M4A, AAC, Ogg/Opus, FLAC, and WebM), and supports free-text prompting, keyterms, and conversation context to steer transcription toward domain vocabulary.
Deepgram: Nova-3
Deepgram Nova-3 general-purpose speech-to-text model with monolingual and multilingual transcription support.
OpenAI: GPT Transcribe
GPT Transcribe is a high-accuracy speech-to-text model from OpenAI. It is suited for recorded audio, streamed file transcription, and committed Realtime turns, with free-form context, keyword hints, and multiple language hints for specialized terms and multilingual speech.

Qwen: Qwen3 ASR 0.6B
Qwen3 ASR 0.6B is a compact automatic speech recognition model from Qwen. It supports multilingual language identification and transcription across 30 languages and 22 Chinese dialects, with streaming and offline inference plus segment-level and word-level timestamps.
Google: Gemini 3.5 Transcribe
Gemini 3.5 Transcribe is a speech-to-text model from Google. It is suited for synchronous transcription that needs word-level timestamps or speaker diarization, with support for up to eight speakers. Audio can be up to one hour, or 30 minutes when timestamps or diarization are enabled.
NVIDIA: Parakeet TDT 0.6B v3
Parakeet TDT 0.6B v3 is NVIDIA's 600M-parameter multilingual speech-to-text model built on the FastConformer-TDT architecture. Trained on the Granary dataset (670,000+ hours of audio), it supports automatic language detection across all official EU languages and achieves a 6.34% average word error rate on the HuggingFace Open ASR Leaderboard. Returns transcribed text with punctuation and segment timestamps.
NVIDIA: Nemotron 3.5 Transcribe
Nemotron 3.5 Transcribe is an open-weight speech recognition model from NVIDIA, a 30B mixture-of-experts with about 4B active parameters. It transcribes recordings of up to one hour in 40 languages, with speaker attribution, speech translation, and prompt-guided vocabulary and style.
OpenAI: GPT-4o Transcribe
GPT-4o Transcribe is OpenAI's high-quality speech-to-text model built on GPT-4o audio capabilities. It's priced per token (input and output), making it suitable for workflows that benefit from token-level billing transparency.

Meta: Muse Voice Transcribe 1.0
Muse Voice Transcribe 1.0 is a synchronous speech-to-text model from Meta. It is suited for push-to-talk, endpointing, and speaker-aware transcription, with keyword biasing for domain terms and language biasing through language-name hints. It accepts mono 16-bit PCM WAV audio at 16 kHz or 24 kHz for recordings up to 10 minutes. It does not provide word-level timestamps or confidence scores, and other audio formats must be converted to WAV before upload.

Mistral: Voxtral Mini Transcribe
Voxtral Mini Transcribe is Mistral's speech-to-text model, derived from the Voxtral Mini family. It accepts audio input and returns transcribed text via the standard transcription API. Suited for transcribing meetings, voice notes, podcasts, and other spoken content.
SpaceXAI: Grok STT 2.0
Grok STT 2.0 is a speech-to-text model from SpaceXAI and the successor to Grok STT 1.0. It is suited for transcription workloads that need word-level timestamps, optional speaker diarization, and multichannel audio.

Mistral: Voxtral Small Transcribe 3
Voxtral Small Transcribe 3 is Mistral's flagship speech-to-text model. It takes audio input and returns a transcript with timestamps and speaker diarization, and supports context biasing for domain-specific vocabulary. Suited for high-accuracy multilingual transcription of meetings, interviews, media, and support calls.

Mistral: Voxtral Mini Transcribe 3
Voxtral Mini Transcribe 3 is Mistral's compact speech-to-text model. It takes audio input and returns a transcript with timestamps and speaker diarization. It also supports context biasing for domain terms. It is small enough to self-deploy in resource-constrained environments, and fits meeting notes, call analytics, and voice-agent pipelines where cost matters.
NVIDIA: Nemotron 3.5 ASR Streaming Multilingual 0.6B
Nemotron 3.5 ASR Streaming Multilingual 0.6B is a speech recognition model from NVIDIA. Its prompt-conditioned, cache-aware FastConformer-RNNT design targets low-latency transcription across more than 40 languages for real-time captioning, voice agents, and multilingual transcription pipelines.
Fish Audio: Transcribe 1 Pro
Transcribe 1 Pro is a speech-to-text model from Fish Audio tuned for interviews, meetings, and podcasts. It labels speakers with inline speaker markers, preserves emotion and vocal-event cues such as [laughter], detects language automatically, and can return timestamped word-level segments.
AssemblyAI: Universal-3.5 Pro
Universal-3.5 Pro is AssemblyAI's speech-to-text model served through its Sync API, returning a complete transcript with word-level timestamps in a single synchronous response for audio up to 20 minutes. It accepts WAV and compressed input (MP3, M4A, AAC, Ogg/Opus, FLAC, and WebM) and supports free-text prompting, keyterms, and conversation context to steer transcription toward domain vocabulary.
ElevenLabs: Scribe v2 Medical
ElevenLabs Scribe v2 Medical is a speech-to-text model tuned for clinical and medical terminology, with word-level timestamps, optional speaker diarization, and audio-event tagging.
Moondream: Parakeet Ultra
Parakeet Ultra is Moondream's post-trained version of NVIDIA's Parakeet TDT 0.6B v3 speech-to-text model, with the same architecture, tokenizer, and 0.6B parameters. It lowers word error rate on the Open ASR Leaderboard English sets (5.80% vs 6.26%) and the 25-language FLEURS set (9.55% vs 11.62%), with larger gains on noisy and long-form audio. Supports 25 European languages and returns transcribed text with segment and word timestamps.
Preguntas frecuentes
Envíe la grabación a POST /api/v1/audio/transcriptions con un slug de modelo de voz a texto de esta página, ya sea como una carga multipart file compatible con OpenAI o como un objeto input_audio que contenga el audio codificado en base64 y su format, como wav o mp3. La respuesta JSON contiene la transcripción en text. Todos los modelos STT de OpenRouter utilizan este endpoint, aunque los formatos de audio y las opciones compatibles varían según el modelo. La guía de voz a texto enumera todos los parámetros de solicitud.
Cada elemento de esta página muestra su precio en la unidad que factura el proveedor. Los modelos basados en la duración cobran por segundo de audio y los modelos basados en tokens cobran por token de entrada y de salida, por lo que la longitud de la transcripción también afecta a su coste. Compare los precios en la misma unidad antes de elegir; no hay una suscripción independiente por proveedor.
Sí. Pase un código language ISO-639-1 como en o ja; si lo omite, el idioma se detecta automáticamente. Los idiomas admitidos varían según el modelo, así que consulte la página del modelo antes de enviar audio en otro idioma.
Sí. Una sola clave de API de OpenRouter funciona con todos los modelos STT de esta página y con todos los demás modelos de OpenRouter. Cree una clave en su página de claves de API.